ECSS Outgassing Screening -- Exposure Execution (space-systems/ecss/q7002-test-execution)
Use when the task is the exposure step of a thermal-vacuum outgassing
screening under ECSS-Q-ST-70-02C: the specimens have been heated under vacuum
against cooled collectors, the run log has been downloaded, and the question
is how much qualified soak time the run delivered and what has to be reported
about the way it delivered it.
Domain quick reference
- Soak time is a conjunction. It accrues only while specimen temperature,
collector temperature and chamber pressure are all inside their limits at
the same moment; two of three is not a partial credit, it is a gap.
- Crediting an interval needs both of its ends. A dropout between two samples
is not interpolated away: the interval before and the interval after the
failing sample both stop counting, which is the conservative and
reproducible reading of the same log.
- Wall-clock elapsed time and qualified soak time are different quantities. A
run that sat in the chamber for twenty-six hours and qualified for sixteen
delivered sixteen, and reporting the twenty-six is how a short soak survives
review.
- A stretch of failing samples is one event. It has a start, an end and the
set of reasons it failed, and the longest such event is the first thing a
reviewer asks about; a finding per sample buries that.
- The phases occur in one order. Stamps that do not increase through
pump-down, heating, stabilisation, soak start, soak end and cool-down
describe a run nobody can reconstruct, whatever the channels say.
Workflow
- Validate the run log: every sample carries a time and three channels, the
pressure is positive, and the time stamps strictly increase.
- Test each sample against the three limits and keep the reasons it failed
rather than a single boolean, so the cause survives into the report.
- Walk the intervals and credit one only when the samples at both ends
qualified, summing the credited intervals into the qualified soak time.
- Group consecutive failing samples into excursions, merging their causes,
and close a stretch that runs to the end of the log.
- Compare the qualified soak time with the required duration, absorbing an
exact equality as representation error rather than by rounding the
requirement.
- Report the qualified time, the elapsed time, every excursion with its
causes and the longest one, and check the phase stamps when they are given.
Pitfalls
- Reporting elapsed time as soak time. They agree only on a run with no
excursion at all, which is exactly the run where the distinction is free.
- Interpolating across a dropout. Nothing is known about the interval a
failed sample sits in, and crediting half of it is an invention that always
favours the run.
- Raising one finding per failing sample. A four-hour excursion sampled every
minute becomes two hundred findings and the reviewer stops reading.
- Dropping the cause when the sample recovers. A soak lost to pressure and a
soak lost to collector control call for different corrective action.
- Reading the phase stamps as paperwork. An out-of-order stamp usually means
the soak clock started before the chamber was stable, and the qualified time
is then being measured against the wrong window.
Behavior contract (gate 3)
The run-log validation, per-sample limit test, interval crediting, excursion
grouping, longest-excursion reporting, phase-order check and the delivered
soak verdict are exercised by the gate 3 contract test:
scripts/test_q7002_test_execution.py against
scripts/q7002_test_execution_logic.py (stdlib unittest, offline). Run:
python3 scripts/test_q7002_test_execution.py
Compliance
- ECSS standards are freely downloadable (ESA); cite the source and
paraphrase per standards-map.yaml.
- compliance: STANDARDS-REF, gated: false.
1---2name: q7002-test-execution3description: "Analyze the run log of a thermal-vacuum outgassing exposure. Use when a soak has finished under ECSS-Q-ST-70-02C and the question is how much of it counts: check that pump-down, heating, stabilisation, soak and cool-down were stamped in the only order they can occur in, credit an interval only when the samples at both ends had specimen temperature, collector temperature and chamber pressure inside their limits together, gather the failing stretches into excursions with their causes, and report qualified soak time beside wall-clock elapsed time. Trigger: ecss, q-st-70-02c, outgassing-exposure-execution, outgassing-qualified-soak-time, outgassing-run-log-excursion, outgassing-exposure-phase-order, outgassing-collector-temperature-hold, outgassing-chamber-pressure-hold."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/q7002-test-execution15 domain: space-systems16 subdomain: ecss17 tags: [ecss, q-st-70-02-outgassing-scope, q7002-test-execution, outgassing-qualified-soak-time, outgassing-run-log-excursion, outgassing-exposure-phase-order, outgassing-collector-temperature-hold, outgassing-chamber-pressure-hold]18 version: 0.1.019 author: Aero Agent Skills20---2122# ECSS Outgassing Screening -- Exposure Execution (space-systems/ecss/q7002-test-execution)2324Use when the task is the exposure step of a thermal-vacuum outgassing25screening under ECSS-Q-ST-70-02C: the specimens have been heated under vacuum26against cooled collectors, the run log has been downloaded, and the question27is how much qualified soak time the run delivered and what has to be reported28about the way it delivered it.2930## Domain quick reference3132- Soak time is a conjunction. It accrues only while specimen temperature,33 collector temperature and chamber pressure are all inside their limits at34 the same moment; two of three is not a partial credit, it is a gap.35- Crediting an interval needs both of its ends. A dropout between two samples36 is not interpolated away: the interval before and the interval after the37 failing sample both stop counting, which is the conservative and38 reproducible reading of the same log.39- Wall-clock elapsed time and qualified soak time are different quantities. A40 run that sat in the chamber for twenty-six hours and qualified for sixteen41 delivered sixteen, and reporting the twenty-six is how a short soak survives42 review.43- A stretch of failing samples is one event. It has a start, an end and the44 set of reasons it failed, and the longest such event is the first thing a45 reviewer asks about; a finding per sample buries that.46- The phases occur in one order. Stamps that do not increase through47 pump-down, heating, stabilisation, soak start, soak end and cool-down48 describe a run nobody can reconstruct, whatever the channels say.4950## Workflow51521. Validate the run log: every sample carries a time and three channels, the53 pressure is positive, and the time stamps strictly increase.542. Test each sample against the three limits and keep the reasons it failed55 rather than a single boolean, so the cause survives into the report.563. Walk the intervals and credit one only when the samples at both ends57 qualified, summing the credited intervals into the qualified soak time.584. Group consecutive failing samples into excursions, merging their causes,59 and close a stretch that runs to the end of the log.605. Compare the qualified soak time with the required duration, absorbing an61 exact equality as representation error rather than by rounding the62 requirement.636. Report the qualified time, the elapsed time, every excursion with its64 causes and the longest one, and check the phase stamps when they are given.6566## Pitfalls6768- Reporting elapsed time as soak time. They agree only on a run with no69 excursion at all, which is exactly the run where the distinction is free.70- Interpolating across a dropout. Nothing is known about the interval a71 failed sample sits in, and crediting half of it is an invention that always72 favours the run.73- Raising one finding per failing sample. A four-hour excursion sampled every74 minute becomes two hundred findings and the reviewer stops reading.75- Dropping the cause when the sample recovers. A soak lost to pressure and a76 soak lost to collector control call for different corrective action.77- Reading the phase stamps as paperwork. An out-of-order stamp usually means78 the soak clock started before the chamber was stable, and the qualified time79 is then being measured against the wrong window.8081## Behavior contract (gate 3)8283The run-log validation, per-sample limit test, interval crediting, excursion84grouping, longest-excursion reporting, phase-order check and the delivered85soak verdict are exercised by the gate 3 contract test:86scripts/test_q7002_test_execution.py against87scripts/q7002_test_execution_logic.py (stdlib unittest, offline). Run:88python3 scripts/test_q7002_test_execution.py8990## Compliance9192- ECSS standards are freely downloadable (ESA); cite the source and93 paraphrase per standards-map.yaml.94- compliance: STANDARDS-REF, gated: false.