ECSS Hybrids — Pre-Seal Burn-In (space-systems/ecss/q6005-pre-seal-burn-in)
Use when the task is clause 10.3.3 of ECSS-Q-ST-60-05: the powered thermal
soak run while the hybrid package is still open, so that whatever it
precipitates can be seen and repaired without breaking a seal. Where the
step sits in the run is graded under the screening sequence; the
thermographic image and the photographic record have leaves of their own.
Domain quick reference
- A soak is worth what it delivers at the reference condition, not what the
chamber log says in hours. A cooler chamber run for the nominal time
delivers less stress, and the Arrhenius conversion is what everyone means
when they call a soak equivalent.
- The activation energy decides how steeply that conversion moves. A
programme that names one uses it; a programme that names none inherits a
documented default rather than an assumption nobody wrote down.
- The band is a band. Too cool and the soak precipitates nothing worth
finding; too hot and it damages the assembly it was run to sort, with no
lid holding anything down.
- Bias is what makes this burn-in rather than a bake. An unbiased soak
precipitates a different and much smaller set of defects, and it is not a
substitute.
- Interruptions accumulate up to a point. A short break leaves the unit hot
and the soak continues; a long one lets it return to ambient, and the
hours before it no longer form part of one continuous soak.
- The chamber band is read over every segment, not only the continuous tail.
A chamber that left the band did so whether or not the soak was later
restarted, and the assembly was there for it.
- The result is the electrical drift across the soak, not the chamber log. A
parameter that moved past its allowance moved because something in the
assembly is unstable, and that unit is what the soak was run to find.
- The package stays open throughout, because the whole point of running the
soak here is that a repair costs nothing to reach.
Workflow
- Take the soak log as segments: hours, chamber temperature, and the break
that followed each one.
- Validate the log — positive hours, a real temperature, a break that is
not negative — and reject a soak with no segment at all.
- Walk the log and restart the accumulation at every break longer than the
permitted interruption, keeping the count of restarts.
- Convert each surviving segment to equivalent hours at the reference
condition, and add them.
- Compare the delivered equivalent hours with the floor the programme asks
for, tolerance included so an exact landing counts.
- Test every declared segment against the permitted chamber band and name
each temperature that left it.
- Grade the soak conditions, overruling a declared duration or band with
what the log actually shows.
- Take each monitored parameter's drift as a fraction of its pre-soak
reading and compare it with the allowance.
- Name the verdict: invalid while a mandatory condition or the index is
short, rejected on a drifting parameter, passed with open actions on a
restart or an unrecorded condition, passed only when nothing is
outstanding.
Pitfalls
- Adding chamber hours and calling the total a soak. Hours at a cooler
chamber are not the hours the requirement is written in.
- Running cool because the assembly is open and fragile, then reporting the
nominal duration. The equivalent hours are what was delivered, and they
are far short.
- Ignoring a long interruption because the total hours still add up. The
unit returned to ambient, and a restarted soak is a shorter soak.
- Grading the chamber band only over the segments that survived the last
restart. The excursion happened to the assembly either way.
- Treating an unbiased bake as a burn-in. It precipitates a different set of
defects and leaves the ones the clause is aimed at in place.
- Measuring only after the soak. Drift is a difference, and without the
pre-soak reading there is nothing to subtract.
- Sealing before the drifting units are dispositioned. The repair was free
while the package was open and is not free afterwards.
Behavior contract (gate 3)
The Arrhenius acceleration factor, equivalent-hours conversion, chamber-band
test, interruption restart rule, delivered-duration sufficiency, parameter
drift fraction and allowance, soak-condition grading, condition index and
burn-in verdict are exercised by the gate 3 contract test:
scripts/test_q6005_pre_seal_burn_in.py against
scripts/q6005_pre_seal_burn_in_logic.py (stdlib unittest, offline). Run:
python3 scripts/test_q6005_pre_seal_burn_in.py
Compliance
- ECSS standards are freely downloadable (ESA); cite the source and
paraphrase per standards-map.yaml.
- compliance: STANDARDS-REF, gated: false.
1---2name: q6005-pre-seal-burn-in3description: "Compute what a biased thermal soak run on an open hybrid microcircuit package actually delivered, and decide whether the unit survived it, under ECSS-Q-ST-60-05 clause 10.3.3. Use when a burn-in log has to be graded rather than counted: convert the chamber hours to equivalent hours at the reference condition through the Arrhenius relation, test every segment against the permitted band, discard the hours before a break long enough to cool the unit, weigh each monitored parameter's drift against its allowance, and return one verdict. Trigger: ecss, q-st-60-05, pre-seal-burn-in, open-package-thermal-soak, burn-in-arrhenius-equivalent-hours, burn-in-interruption-restart, pre-and-post-soak-parameter-drift, pre-seal-burn-in-verdict."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/q6005-pre-seal-burn-in15 domain: space-systems16 subdomain: ecss17 tags: [ecss, q-st-60-hybrid-scope, q-st-60-05, q6005-pre-seal-burn-in, open-package-thermal-soak, burn-in-arrhenius-equivalent-hours, burn-in-interruption-restart, pre-and-post-soak-parameter-drift, pre-seal-burn-in-verdict]18 version: 0.1.019 author: Aero Agent Skills20---2122# ECSS Hybrids — Pre-Seal Burn-In (space-systems/ecss/q6005-pre-seal-burn-in)2324Use when the task is clause 10.3.3 of ECSS-Q-ST-60-05: the powered thermal25soak run while the hybrid package is still open, so that whatever it26precipitates can be seen and repaired without breaking a seal. Where the27step sits in the run is graded under the screening sequence; the28thermographic image and the photographic record have leaves of their own.2930## Domain quick reference3132- A soak is worth what it delivers at the reference condition, not what the33 chamber log says in hours. A cooler chamber run for the nominal time34 delivers less stress, and the Arrhenius conversion is what everyone means35 when they call a soak equivalent.36- The activation energy decides how steeply that conversion moves. A37 programme that names one uses it; a programme that names none inherits a38 documented default rather than an assumption nobody wrote down.39- The band is a band. Too cool and the soak precipitates nothing worth40 finding; too hot and it damages the assembly it was run to sort, with no41 lid holding anything down.42- Bias is what makes this burn-in rather than a bake. An unbiased soak43 precipitates a different and much smaller set of defects, and it is not a44 substitute.45- Interruptions accumulate up to a point. A short break leaves the unit hot46 and the soak continues; a long one lets it return to ambient, and the47 hours before it no longer form part of one continuous soak.48- The chamber band is read over every segment, not only the continuous tail.49 A chamber that left the band did so whether or not the soak was later50 restarted, and the assembly was there for it.51- The result is the electrical drift across the soak, not the chamber log. A52 parameter that moved past its allowance moved because something in the53 assembly is unstable, and that unit is what the soak was run to find.54- The package stays open throughout, because the whole point of running the55 soak here is that a repair costs nothing to reach.5657## Workflow58591. Take the soak log as segments: hours, chamber temperature, and the break60 that followed each one.612. Validate the log — positive hours, a real temperature, a break that is62 not negative — and reject a soak with no segment at all.633. Walk the log and restart the accumulation at every break longer than the64 permitted interruption, keeping the count of restarts.654. Convert each surviving segment to equivalent hours at the reference66 condition, and add them.675. Compare the delivered equivalent hours with the floor the programme asks68 for, tolerance included so an exact landing counts.696. Test every declared segment against the permitted chamber band and name70 each temperature that left it.717. Grade the soak conditions, overruling a declared duration or band with72 what the log actually shows.738. Take each monitored parameter's drift as a fraction of its pre-soak74 reading and compare it with the allowance.759. Name the verdict: invalid while a mandatory condition or the index is76 short, rejected on a drifting parameter, passed with open actions on a77 restart or an unrecorded condition, passed only when nothing is78 outstanding.7980## Pitfalls8182- Adding chamber hours and calling the total a soak. Hours at a cooler83 chamber are not the hours the requirement is written in.84- Running cool because the assembly is open and fragile, then reporting the85 nominal duration. The equivalent hours are what was delivered, and they86 are far short.87- Ignoring a long interruption because the total hours still add up. The88 unit returned to ambient, and a restarted soak is a shorter soak.89- Grading the chamber band only over the segments that survived the last90 restart. The excursion happened to the assembly either way.91- Treating an unbiased bake as a burn-in. It precipitates a different set of92 defects and leaves the ones the clause is aimed at in place.93- Measuring only after the soak. Drift is a difference, and without the94 pre-soak reading there is nothing to subtract.95- Sealing before the drifting units are dispositioned. The repair was free96 while the package was open and is not free afterwards.9798## Behavior contract (gate 3)99100The Arrhenius acceleration factor, equivalent-hours conversion, chamber-band101test, interruption restart rule, delivered-duration sufficiency, parameter102drift fraction and allowance, soak-condition grading, condition index and103burn-in verdict are exercised by the gate 3 contract test:104scripts/test_q6005_pre_seal_burn_in.py against105scripts/q6005_pre_seal_burn_in_logic.py (stdlib unittest, offline). Run:106python3 scripts/test_q6005_pre_seal_burn_in.py107108## Compliance109110- ECSS standards are freely downloadable (ESA); cite the source and111 paraphrase per standards-map.yaml.112- compliance: STANDARDS-REF, gated: false.