Space Product Assurance: Worst Case Analysis
Space Product Assurance: Worst Case Analysis
14 January 2011
Space product
assurance
Worst case analysis
ECSS Secretariat
ESA-ESTEC
Requirements & Standards Division
Noordwijk, The Netherlands
ECSS‐Q‐HB‐30‐01A
14 January 2011
Foreword
This Handbook is one document of the series of ECSS Documents intended to be used as supporting
material for ECSS Standards in space projects and applications. ECSS is a cooperative effort of the
European Space Agency, national space agencies and European industry associations for the purpose
of developing and maintaining common standards.
This handbook has been prepared by the ECSS‐Q‐HB‐30‐01 Working Group, reviewed by the ECSS
Executive Secretariat and approved by the ECSS Technical Authority.
Disclaimer
ECSS does not provide any warranty whatsoever, whether expressed, implied, or statutory, including,
but not limited to, any warranty of merchantability or fitness for a particular purpose or any warranty
that the contents of the item are error‐free. In no respect shall ECSS incur any liability for any
damages, including, but not limited to, direct, indirect, special, or consequential damages arising out
of, resulting from, or in any way connected to the use of this document, whether or not based upon
warranty, business agreement, tort, or otherwise; whether or not injury was sustained by persons or
property or otherwise; and whether or not loss was sustained from, or arose out of, the results of, the
item, or any services that may be provided by ECSS.
Published by: ESA Requirements and Standards Division
ESTEC, P.O. Box 299,
2200 AG Noordwijk
The Netherlands
Copyright: 2011© by the European Space Agency for the members of ECSS
2
ECSS‐Q‐HB‐30‐01A
14 January 2011
Change log
ECSS‐Q‐HB‐30‐01A First issue
14 January 2011
3
ECSS‐Q‐HB‐30‐01A
14 January 2011
Table of contents
1 Scope.......................................................................................................................6
2 References ..............................................................................................................7
4 General methodology...........................................................................................11
4.1 Introduction............................................................................................................... 11
4.2 Flow diagram of WCA .............................................................................................. 11
4.3 Identification of the critical aspects w.r.t. worst case performance........................... 13
4.4 Evaluation of worst case performance ..................................................................... 13
4.5 Comparison of WCA with requirements ................................................................... 14
4
ECSS‐Q‐HB‐30‐01A
14 January 2011
5 Analysis parameters and technical issues.........................................................15
5.1 Definition of worst case parameters within parts database ...................................... 15
5.1.1 Variation factors.......................................................................................... 15
5.1.2 Summary on deviations .............................................................................. 19
5.2 Phase and timing considerations within the WCA .................................................... 20
5.2.1 Introduction................................................................................................. 20
5.2.2 Timing of transient pulses........................................................................... 20
5.3 Numerical analysis techniques ................................................................................. 20
5.3.1 Approach .................................................................................................... 20
5.3.2 Extreme value analysis............................................................................... 21
5.3.3 Extreme value analysis combined approach .............................................. 21
5.3.4 Root-sum-squared analysis........................................................................ 21
5.3.5 Monte Carlo analysis .................................................................................. 21
Figures
Figure4-1: Flow diagram of WCA ........................................................................................... 12
Tables
Table 5-1: Deviations and attributes summary....................................................................... 19
Table 5-2Numerical techniques and value summary ............................................................. 22
5
ECSS‐Q‐HB‐30‐01A
14 January 2011
1
Scope
This handbook provides guidelines to perform the worst case analysis. It applies to all electrical and
electronic equipment. This worst case analysis (WCA) method can also be applied at subsystem level
to justify electrical interface specifications and design margins for equipment. It applies to all project
phases where electrical interface requirements are established and circuit design is carried out.
The worst case analysis is generally carried out when designing the circuit. For selected circuitry,
worst case analysis (WCA) can be used to validate a conceptual design approach.
6
ECSS‐Q‐HB‐30‐01A
14 January 2011
2
References
ECSS‐ST‐00‐01 ECSS system ‐ Glossary of terms
ECSS‐E‐ST‐10‐02 Space engineering ‐Verification
ECSS‐Q‐ST‐30 Space product assurance ‐ Dependability
ECSS‐Q‐ST‐30‐11 Space product assurance ‐ Derating ‐ EEE
components
ECSS‐Q‐TM‐30‐12 Space product assurance – End‐of‐life parameters
drifts ‐ EEE components
ECSS‐Q‐ST‐30‐02 Space product assurance ‐ Failure modes, effects and
criticality analysis
ECSS‐Q‐ST‐40‐02 Space product assurance ‐ Hazard analysis
ECSS‐Q‐TM‐40‐04 Space product assurance ‐ Sneak analysis
ECSS‐Q‐ST‐40‐12 Space product assurance ‐ Fault tree analysis –
Adoption notice ECSS / IEC61025
CRTAWCCA Worst Case Circuit Analysis Application Guidelines,
1993 Reliability Analysis Center, Rome NY, U.S.A
JPL D‐5703 Jet Propulsion Laboratory Reliability Analyses
Handbook
7
ECSS‐Q‐HB‐30‐01A
14 January 2011
3
Terms, definitions and abbreviated terms
8
ECSS‐Q‐HB‐30‐01A
14 January 2011
3.2.8 radiation
phenomenon by which energy, in form of waves or particles, emanates from a source into space
Example Trapped electrons, trapped protons and solar protons.
9
ECSS‐Q‐HB‐30‐01A
14 January 2011
10
ECSS‐Q‐HB‐30‐01A
14 January 2011
4
General methodology
4.1 Introduction
The worst case analysis (WCA) is performed on electronic and electrical equipment to demonstrate
that it performs within specification despite particular variations in its constituent part parameters
and the imposed environment, at the end of overall lifetime (EOL).
A good survey of worst case circuit analysis can be found in CRTAWCCA “Worst Case Circuit
Analysis Application Guidelines, 1993 Reliability Analysis Center, Rome NY, U.S.A.”.
11
ECSS‐Q‐HB‐30‐01A
14 January 2011
Figure 4‐1: Flow diagram of WCA
12
ECSS‐Q‐HB‐30‐01A
14 January 2011
13
ECSS‐Q‐HB‐30‐01A
14 January 2011
power supply drifts). Otherwise the WCA should justify the absence of such effects, or should
quantify such effects as negligible, in comparison to other error sources.
A combination of testing and analysis may be employed to obtain results through actual
measurements.
14
ECSS‐Q‐HB‐30‐01A
14 January 2011
5
Analysis parameters and technical issues
5.1.1.1 General
For each physical parameter of the component affecting the worst case parameter analysed, the value
of the variation linked to each environment or interface stimulus should be determined. Sources of
variation include:
the reference value (e.g. typical and adjusted),
the initial tolerance,
sensitivity to electrical interfaces (e.g. power supply, shared mode on inputs and output loads),
and
sensitivity to ageing and environmental conditions (e.g. temperature, radiation and EMC).
5.1.1.3 Compensation
If the circuit compensates initial tolerance or environmental variations (such as temperature) the
analysis report should include a justification for the residual variation.
5.1.1.4 Radiation
15
ECSS‐Q‐HB‐30‐01A
14 January 2011
These parametric changes are related to the influence of the accumulated total dose radiation received
throughout the mission lifetime.
16
ECSS‐Q‐HB‐30‐01A
14 January 2011
5.1.1.4.5 Single event effects
Single event effects (SEE) are anomalies and thus not a variation factor to be considered in the scope of
the WCA. If the SEE assessment of the circuit requires countermeasures for SEE, these should be
described in a design justification document. If, as a result of SEE assessment, a protection circuit is
included to prevent circuit failures or degradation during SEE events, certainly for nominal operation,
this circuit should be considered in the WCA.
5.1.1.5 Temperature
a. Parameter variations due to temperature variations are applicable to all passive and active
components.
b. The parameter variations are taken from the component specification, manufacturer’s
specifications, data sheet or test results. These parameter variations generally have biased
values and are expressed as a delta per degree Celsius (in % or in value) with respect to the
value within the component specification at the reference condition. There can also be a random
part with respect to the bias value. For passive parts the parameter variation over temperature
is usually a fixed value.
c. Often the component specification does not contain the parameter variation in the necessary
parametric form required for the evaluation of the worst case performance of the circuit. In this
case, the parameter variations should be derived considering, for instance, measurement data
and component physics.
d. In the component specification these variations generally apply over the complete temperature
range of the component (‐55 C to +125 C). The thermal analysis of the equipment is provided
as an input with the actual temperature range of each component, which is usually lower than
this complete temperature range. The thermal analysis considers the temperature rise between
the PCB temperature, the case temperature and the junction temperature.
e. Within the WCA the temperature of the component is the ambient temperature when the
equipment thermal interface varies over the acceptance temperature range. In the worst case
analysis, the minimum temperature of each component is the minimum acceptance
environmental temperature of the equipment. The maximum temperature is the temperature
determined within the thermal analysis for the extreme acceptance temperature of the
equipment. As long as the results from the thermal analysis are not available, the component
can be assumed to operate at the maximum operating temperature.
5.1.1.7 Ageing
5.1.1.7.1 Introduction
Parameter variations due to ageing concern all passive and all active components.
This variation is a function of time and temperature, junction temperature for active components, and
case temperature for passive components.
17
ECSS‐Q‐HB‐30‐01A
14 January 2011
The ageing effects are specific to each family of components.
The variations are taken from the component specifications, life test data or worst case parameter
database.
5.1.1.7.2 Extrapolation of ageing data from the lifetime assumed in database to the
design lifetime
If the design lifetime is different from the lifetime assumed in the database, the effective ageing data
should be extrapolated from the data available in the database:
The linear extrapolation (conservative approach) should be applied. The use of other extrapolations
may be adopted with adequate justification.
For a design lifetime shorter than the database, the database value for the next longer time interval
should be assumed. The parameter variation data should not be interpolated between database values
without justification, as the ageing process cannot be assumed a priori to be linear.
5.1.1.7.3 Extrapolation of ageing data from the temperature assumed in the database to
the maximum temperature of the component in the application
To extrapolate ageing data from the temperature assumed in the database to the maximum
temperature of the component for the same duration, the law of Arrhenius should be used:
E 1 1
q 2 q1 exp A
k T1 T2
where
q1 and q2 are the parameters at temperature T1 ant T2, respectively;
T1 and T2 are the temperatures in Kelvin at which q1 and q2 are measured,
respectively;
EA is the activation energy in eV;
k is the Boltzmann constant (8,62 10‐5 eV/K);
exp(x) = ex, where e is the base of the natural logarithm.
The use of other extrapolations may be adopted with adequate justification.
Some typical activation energy values are (these are default values; with the appropriate justification
and reference, other values can be used):
Semiconductors
GaAs 1,4 eV
Silicon 1,1 eV
Resistances
Metal film, thin film 1,35 eV
Carbon 1 eV
Wirewound 1 eV
Capacitances
Ceramic 1,67 eV
Porcelain, glass, mica 1,1 eV
18
ECSS‐Q‐HB‐30‐01A
14 January 2011
Film, plastic 3,4 eV
Tantalum 0,43 eV
NOTE For a mission duration of 10 years, we want the drift in ageing of
VZ at T2 = 85 C. We take an activation energy of 1,1 eV
(semiconductors).
We know that q1 = 2 % at T1 = 110 C, what is the value of q2 ?
1,1 1 1
q 2 2 exp 5 => q 2 0,195%
8,62 10 273 110 273 85
Table 5‐1: Deviations and attributes summary
Initial tolerance Ageing Temperature Radiation EMC and variation
of electrical signals
Components All All All Active only All
concerned
Type Random Biased (sometimes Biased Biased Biased (sometimes
random) (sometimes random)
random)
Function of Intrinsic Time, Temperature Dose received Design
temperature range
Where to find Component Worst case Component Component Interface
the data? specification parameter specification specification specification and
database radiation tests EMC requirements
Specific case Compensation ‐ Temperature ‐ ‐
by alignment compensated
circuit
19
ECSS‐Q‐HB‐30‐01A
14 January 2011
5.2.1 Introduction
This clause provides inputs to take into account the phase and timing problems. Some specific
problems are detailed in this clause. Nevertheless, other timing problems (such as reset duration and
voltage level) should be analysed in the WCA.
Within the WCA it should also be demonstrated that the timing conditions of the signals are such that
the circuit operates properly together under simultaneous worst case source and load conditions. It is
suggested that particular attention be paid to noise margin at the interface when performing the worst
case analysis.
5.2.2.1 General
All sequential circuits should have a worst case timing diagram made to determine the effects of
variation in switching times of the installed circuits. There are many factors that affect timing. They
include supply voltage, capacitive loading, clock or oscillator instability, and slope of clock rising and
falling edge.
5.3.1 Approach
The worst case analysis can be performed using four different approaches:
extreme value analysis;
extreme value analysis combined approach;
root‐sum‐squared analysis;
Monte Carlo analysis.
20
ECSS‐Q‐HB‐30‐01A
14 January 2011
d n , a
2
n
d(n, ) = 1,36 for = 0,05 (confidence level = 95%)
d(n, ) = 1,68 for = 0,01 (confidence level = 99 %)
The advantages of the approaches are shown in Table 5‐2
21
ECSS‐Q‐HB‐30‐01A
14 January 2011
Table 5‐2: Numerical techniques and value summary
Advantages Disadvantages
EVA Provides most readily obtainable Results in pessimistic estimate of
estimate of worst case performance. circuit worst case performance.
Does not require statistical inputs for If circuit fails, there is insufficient data
circuit parameters. to assess risk.
Database need only provide part
parameter variation extremes.
EVA Results in more realistic estimate of Valid for standard deviations
combined worst case performance than EVA. (moment methods) but not always
approach This method can be implemented valid for maximum values.
with simple calculations. Requires accurate knowledge of piece
part parameter PDF.
It is strictly valid only for Gaussian
variables.
Risk of over‐dimensioning is as high
as for EVA.
RSS Results in more realistic estimate of The computation of the standard
worst case performance than EVA. deviation of piece part parameter’s
Knowledge of parameter PDF not probability distribution.
required. Assumes circuit sensitivities remain
Provides some degree of risk constant over range of parameter
assessment in terms of percentage of variability.
units to pass or fail. Uses approximation: circuit
performance variability is normally
distributed (central limit theorem).
Monte Provides the most realistic estimate of Consumes large amount of CPU time.
Carlo true worst case performance of the
three methods.
Provides additional information
which can be applied to risk
assessment.
For selection of the appropriate method for the first three techniques, see subclause 5.2.3 of
CRTAWCCA (Worst Case Circuit Analysis Application Guidelines, 1993 Reliability Analysis Center,
Rome NY, U.S.A.).
22
ECSS‐Q‐HB‐30‐01A
14 January 2011
6
WCA and project phases
A detailed WCA during the design phase can be used to find design problems that were not found
during the test phase due to temperature extremes, age or radiation.
For selected circuitry, a preliminary WCA should be available to validate a conceptual design
approach at PDR.
The assumptions and approach to be used in the analyses should be checked against reliability
assessments prior to the performance of the analyses.
The critical aspects with respect to worst case performance and the critical circuit parameters should
be identified. These critical aspects can be identified from results of other analyses (e.g. FMECA, sneak
analysis). The circuits models and equations should be defined. If a nonconformance is identified, a
redesign of the circuit can be proposed to achieve conformance or an optimization of the analysis can
be investigated.
If design changes are made, either as a result of the WCA or for other reasons, the WCA should be
updated using the new circuit.
Results of such an analysis are generally presented in the frame of the circuit CDR.
23