ECSS Space Environment — SEE Hardness Assurance (space-systems/ecss/e1012-see-hardness)
Use when the task is the §9.5 single-event effects hardness assurance
procedure of ECSS-E-ST-10-12C — determining each electronic device's
susceptibility to ion-induced and proton/neutron-induced SEE, predicting
the on-orbit rate, comparing it against the system error-rate requirement,
and assigning the hardness assurance category that governs acceptance
testing and lot-screening obligations.
Domain quick reference
- §9.5 organises the SEE hardness assurance work around three particle
environments and two effect families. Ion-induced effects are driven by
the heavy-ion LET spectrum; proton/neutron-induced effects are driven by
the energy spectrum of trapped and solar protons plus secondary neutrons.
The two rate contributions are summed for comparison against the
system-level error-rate requirement.
- Effect types split into destructive (SEL — single-event latchup; SEB —
single-event burnout; SEGR — single-event gate rupture) and
non-destructive (SEU — single-event upset; SEFI — single-event functional
interrupt; SETR — single-event transient; SEHE — single-event hard error).
Destructive effects require the most stringent hardness assurance controls
regardless of whether the predicted rate meets the numerical requirement.
- Device susceptibility is determined by the LET threshold (ions) and the
proton energy threshold. A device whose ion LET threshold is at or above
the ion immunity level (37 MeV·cm²/mg) needs no ion hardness assurance
measures. A device whose proton threshold is at or above the proton
immunity level (200 MeV) needs no proton/neutron measures.
- The SEE rate is estimated from the Weibull cross-section model: the device
cross-section as a function of LET (or energy) is integrated against the
mission particle spectrum, yielding a predicted rate in events per device
per day. The Weibull model requires four parameters: saturation cross-
section (σ_sat), onset threshold (LET₀ or E₀), width (W), and shape (s).
- Three hardness assurance categories govern acceptance obligations.
HA1 applies to any device susceptible to a destructive SEE effect: lot
screening plus acceptance test every lot. HA2 applies to a device with a
non-destructive SEE effect whose predicted rate exceeds the requirement:
representative sample test. HA3 applies to a device whose predicted rate
is within the requirement: compliant by analysis, with the analysis
documented and reviewed when the environment estimate is updated.
Workflow
- Identify every electronic device in the design and enumerate the SEE
effect types relevant to each device (SEU, SEL, SEB, SEFI, SETR, SEHE,
SEGR). Reject an unrecognized effect type before it enters the rate
calculation.
- For each device-effect pair, obtain the device cross-section parameters
(σ_sat, LET threshold or energy threshold, Weibull width W, shape s)
from the device test data or manufacturer data sheet. If data is
unavailable, apply a conservative bounding assumption and document it.
- Check ion immunity: if the LET threshold is at or above
37 MeV·cm²/mg, record the device as ion-immune for this effect and skip
the ion rate step.
- If not ion-immune, integrate the Weibull ion cross-section against the
mission heavy-ion LET spectrum to obtain the ion-induced SEE rate
(events/device/day): rate_ion = Σ σ_Weibull(LET_i) × flux_i over all
LET bins.
- Check proton immunity: if the proton energy threshold is at or above
200 MeV, record the device as proton-immune for this effect and skip
the proton/neutron rate step.
- If not proton-immune, integrate the Weibull proton cross-section against
the mission proton/neutron energy spectrum to obtain the proton/neutron-
induced SEE rate: rate_pn = Σ σ_Weibull(E_i) × flux_i over all energy
bins.
- Sum the ion and proton/neutron rates to obtain the total predicted rate:
rate_total = rate_ion + rate_pn.
- Compare rate_total against the system error-rate requirement for the
device-effect pair. Flag a violation when rate_total exceeds the
requirement.
- Assign the hardness assurance category: HA1 if the effect is
destructive; HA2 if non-destructive and the rate violates the
requirement; HA3 if non-destructive and the rate is within the
requirement.
- Aggregate per-device findings. A device has no outstanding findings only
when all effect assessments are free of rate violations (HA1 devices
still receive HA1 regardless of their numerical rate).
Pitfalls
- Treating a destructive-effect device as compliant when its predicted rate
numerically meets the requirement — HA1 is mandatory for destructive
effects regardless of the numerical comparison; lot screening cannot be
waived by rate analysis alone.
- Summing ion and proton rates from mismatched environments — the ion LET
spectrum and the proton/neutron energy spectrum must both derive from the
same mission orbit and duration, and both must use the same shielding
depth applied to the device location.
- Ignoring the immunity thresholds — applying the Weibull integration to a
device that sits above the ion immunity LET threshold overstates risk and
may trigger unnecessary HA1 or HA2 obligations; conversely, failing to
check immunity and skipping the integration for a sensitive device
understates risk.
- Using a manufacturer's nominal Weibull fit without applying a margin to
the parameters — test data scatter and lot-to-lot variability mean that
the nominal fit understates the cross-section at the worst-case end; the
assessment should reflect whether a margin factor has been applied.
- Closing the review without updating the rate assessment when the mission
orbit or shielding design changes — the §9.5 procedure is iterated every
time an environment input or shielding assumption is revised.
Behavior contract (gate 3)
The SEE-type categorization, Weibull cross-section evaluation, ion and
proton/neutron rate integration, hardness assurance category assignment, and
full device review logic are exercised by the gate 3 contract test:
scripts/test_e1012_see_hardness.py against
scripts/e1012_see_hardness_logic.py (stdlib unittest, offline). Run:
python3 scripts/test_e1012_see_hardness.py
Compliance
- ECSS standards are freely downloadable (ESA); cite the source and
paraphrase per standards-map.yaml.
- compliance: STANDARDS-REF, gated: false.
1---2name: e1012-see-hardness3description: "Use when run SEE hardness assurance for a space electronic device under ECSS-E-ST-10-12C §9.5: determine the device's susceptibility to ion-induced and proton/neutron-induced single-event effects, predict the on-orbit SEE rate using device cross-section parameters and the mission particle spectrum, compare the predicted rate against the allowed-error-rate requirement, and assign the hardness assurance category that sets acceptance testing and lot-screening obligations. Trigger: ecss, e-st-10-system-scope, see, single-event-effect, seu, sel, seb, see-rate, hardness-assurance, ion-environment, proton-neutron."4license: Apache-2.05compliance: STANDARDS-REF6standards:7 - id: ecss8 reference-only: true9gated: false10domain: space-systems11pack: space-systems12compatibility: "agentskills.io SKILL.md; any SKILL.md host (Claude Code, Hermes, OpenClaw)"13metadata:14 source: https://skillmd.com/skills/ashfordeou/e1012-see-hardness15 domain: space-systems16 subdomain: ecss17 tags: [ecss, e-st-10-system-scope, see, single-event-effect, seu, sel, seb, hardness-assurance, ion-see-rate, proton-see-rate]18 version: 0.1.019 author: Aero Agent Skills20---2122# ECSS Space Environment — SEE Hardness Assurance (space-systems/ecss/e1012-see-hardness)2324Use when the task is the §9.5 single-event effects hardness assurance25procedure of ECSS-E-ST-10-12C — determining each electronic device's26susceptibility to ion-induced and proton/neutron-induced SEE, predicting27the on-orbit rate, comparing it against the system error-rate requirement,28and assigning the hardness assurance category that governs acceptance29testing and lot-screening obligations.3031## Domain quick reference3233- §9.5 organises the SEE hardness assurance work around three particle34 environments and two effect families. Ion-induced effects are driven by35 the heavy-ion LET spectrum; proton/neutron-induced effects are driven by36 the energy spectrum of trapped and solar protons plus secondary neutrons.37 The two rate contributions are summed for comparison against the38 system-level error-rate requirement.39- Effect types split into destructive (SEL — single-event latchup; SEB —40 single-event burnout; SEGR — single-event gate rupture) and41 non-destructive (SEU — single-event upset; SEFI — single-event functional42 interrupt; SETR — single-event transient; SEHE — single-event hard error).43 Destructive effects require the most stringent hardness assurance controls44 regardless of whether the predicted rate meets the numerical requirement.45- Device susceptibility is determined by the LET threshold (ions) and the46 proton energy threshold. A device whose ion LET threshold is at or above47 the ion immunity level (37 MeV·cm²/mg) needs no ion hardness assurance48 measures. A device whose proton threshold is at or above the proton49 immunity level (200 MeV) needs no proton/neutron measures.50- The SEE rate is estimated from the Weibull cross-section model: the device51 cross-section as a function of LET (or energy) is integrated against the52 mission particle spectrum, yielding a predicted rate in events per device53 per day. The Weibull model requires four parameters: saturation cross-54 section (σ_sat), onset threshold (LET₀ or E₀), width (W), and shape (s).55- Three hardness assurance categories govern acceptance obligations.56 HA1 applies to any device susceptible to a destructive SEE effect: lot57 screening plus acceptance test every lot. HA2 applies to a device with a58 non-destructive SEE effect whose predicted rate exceeds the requirement:59 representative sample test. HA3 applies to a device whose predicted rate60 is within the requirement: compliant by analysis, with the analysis61 documented and reviewed when the environment estimate is updated.6263## Workflow64651. Identify every electronic device in the design and enumerate the SEE66 effect types relevant to each device (SEU, SEL, SEB, SEFI, SETR, SEHE,67 SEGR). Reject an unrecognized effect type before it enters the rate68 calculation.692. For each device-effect pair, obtain the device cross-section parameters70 (σ_sat, LET threshold or energy threshold, Weibull width W, shape s)71 from the device test data or manufacturer data sheet. If data is72 unavailable, apply a conservative bounding assumption and document it.733. Check ion immunity: if the LET threshold is at or above74 37 MeV·cm²/mg, record the device as ion-immune for this effect and skip75 the ion rate step.764. If not ion-immune, integrate the Weibull ion cross-section against the77 mission heavy-ion LET spectrum to obtain the ion-induced SEE rate78 (events/device/day): rate_ion = Σ σ_Weibull(LET_i) × flux_i over all79 LET bins.805. Check proton immunity: if the proton energy threshold is at or above81 200 MeV, record the device as proton-immune for this effect and skip82 the proton/neutron rate step.836. If not proton-immune, integrate the Weibull proton cross-section against84 the mission proton/neutron energy spectrum to obtain the proton/neutron-85 induced SEE rate: rate_pn = Σ σ_Weibull(E_i) × flux_i over all energy86 bins.877. Sum the ion and proton/neutron rates to obtain the total predicted rate:88 rate_total = rate_ion + rate_pn.898. Compare rate_total against the system error-rate requirement for the90 device-effect pair. Flag a violation when rate_total exceeds the91 requirement.929. Assign the hardness assurance category: HA1 if the effect is93 destructive; HA2 if non-destructive and the rate violates the94 requirement; HA3 if non-destructive and the rate is within the95 requirement.9610. Aggregate per-device findings. A device has no outstanding findings only97 when all effect assessments are free of rate violations (HA1 devices98 still receive HA1 regardless of their numerical rate).99100## Pitfalls101102- Treating a destructive-effect device as compliant when its predicted rate103 numerically meets the requirement — HA1 is mandatory for destructive104 effects regardless of the numerical comparison; lot screening cannot be105 waived by rate analysis alone.106- Summing ion and proton rates from mismatched environments — the ion LET107 spectrum and the proton/neutron energy spectrum must both derive from the108 same mission orbit and duration, and both must use the same shielding109 depth applied to the device location.110- Ignoring the immunity thresholds — applying the Weibull integration to a111 device that sits above the ion immunity LET threshold overstates risk and112 may trigger unnecessary HA1 or HA2 obligations; conversely, failing to113 check immunity and skipping the integration for a sensitive device114 understates risk.115- Using a manufacturer's nominal Weibull fit without applying a margin to116 the parameters — test data scatter and lot-to-lot variability mean that117 the nominal fit understates the cross-section at the worst-case end; the118 assessment should reflect whether a margin factor has been applied.119- Closing the review without updating the rate assessment when the mission120 orbit or shielding design changes — the §9.5 procedure is iterated every121 time an environment input or shielding assumption is revised.122123## Behavior contract (gate 3)124125The SEE-type categorization, Weibull cross-section evaluation, ion and126proton/neutron rate integration, hardness assurance category assignment, and127full device review logic are exercised by the gate 3 contract test:128scripts/test_e1012_see_hardness.py against129scripts/e1012_see_hardness_logic.py (stdlib unittest, offline). Run:130python3 scripts/test_e1012_see_hardness.py131132## Compliance133134- ECSS standards are freely downloadable (ESA); cite the source and135 paraphrase per standards-map.yaml.136- compliance: STANDARDS-REF, gated: false.