Skip to content

Firmware 2.2.0: driver status check; driver output off during a thermal lockout - #35

Merged
daniel-frenkel merged 4 commits into
mainfrom
fw/driver-check-2.2.0
Nov 17, 2026
Merged

daniel-frenkel merged 4 commits into
mainfrom
fw/driver-check-2.2.0

Conversation

@daniel-frenkel

@daniel-frenkel daniel-frenkel commented Nov 15, 2026 •

Copy link
Copy Markdown
Member

Brings the driver status check into 2.2.0.

  • Every 500 ms the unit reads the driver's status flags. If the driver reports a reset, an undervoltage or a fault, the unit:
    • stops and switches the driver output off;
    • re-applies the driver settings;
    • switches the output back on once the settings read back as written.
  • During a thermal lockout the driver output is off. It comes back on only after the lockout clears and the settings are confirmed.
  • While the settings are being re-applied, Open, Close, Set position and Auto-Calibrate are refused, with the log line "driver held while its settings are re-applied". STOP always works.
  • At boot, the status flags are cleared after the driver is configured.
  • After such a stop:
    • the next full Close may run up to 1.3 mm further past zero, and the next full Open (with a learned stroke) up to 1.3 mm further, so each still ends on its switch;
    • S4 or S5 ends this, and so does a restart.
  • The bench build logs the count at each S5 trip.

This replaces #32 for 2.2.0. #32's two commits are cherry-picked onto main, plus one commit for 2.2.0 and one for the next-move allowance.

Checks:

  • The product, bench and TP-1 configs are valid, and the product builds.
  • The entity gate against main is clean.
  • The behaviour gate lists the changed and new actions.

🤖 Generated with Claude Code

daniel-frenkel and others added 4 commits November 14, 2026 18:47
…kout

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Cherry-picked from e8e3b53; the changelog entry sits under 2.2.0.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…sition

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… to their switch

The next full Close may run 1.3 mm further past zero, and the next full
Open with a learned stroke 1.3 mm further, so each still ends on its
switch; S4 or S5 clears it, and a restart clears it. The bench build logs
the count at each S5 trip.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@daniel-frenkel
daniel-frenkel marked this pull request as ready for review November 17, 2026 05:07
@daniel-frenkel
daniel-frenkel merged commit f3643ab into main Nov 17, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant