Auto-resume only on NOT FEEDING; purge check fails on pressure left after the purge

A CLOG DETECTED pause is usually a real blockage: on 2026-11-17 a nozzle
blocked right after a swap, the purge check passed (a blocked nozzle also
builds pressure) and the automatic resume only printed air until the next
CLOG pause. CLOG messages are still cleared after a resume, so a leftover
one cannot hide the next NOT FEEDING message.

The purge check now also fails when bd_pressure's delta is still above
pressure_max_delta (15) when the check stops: 16 good purges -3..+10, two
long purges into the blocked nozzle +19/+21 (the short ones did not show it).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
mangavalkandClaude Opus 5.5 committed 2026-11-17 11:16:50 +01:00
1 parent 385adb5283
commit 2629e70fad
4 files changed
+27 -8

No files matched your search

+1 -1
View File
@@ -39,7 +39,7 @@ everything below, grouped by topic. Comment out what you do not use there.
| `lane_tools.cfg` | `LANE_FEED_UNTIL_HUB_OPEN`, `LANE_FEED_ABORT` | AFC |
| `purge_check.cfg` | `BLOBIFIER_CHECKED`: checks every AFC purge with bd_pressure (and optionally bdwidth motion); a bd_pressure that does not answer fails the check, `PURGE_CHECK_IGNORE_SENSOR` lets the current print finish without it; when the previous filament (saved `nozzle_last_temp`) needs more than 10°C above the new one, the purge runs at the previous filament's temperature and the nozzle cools back afterwards, so its leftover melts out | Blobifier, AFC, `bdpressure_check.py` from the bd_pressure repo + its load-check firmware, `extras/bdwidth_status.py` |
| `flex_lanes.cfg` | Lanes holding TPU/TPE (material from Spoolman) get slower long moves (50 mm/s, AFC's minimum) through the tube; back to the configured speed when the lane holds something else. Polls every 5 s, reapplies after a restart. The last stretch into the extruder (`tool_load_speed`) stays as configured | AFC |
| `auto_resume.cfg` | After an AFC buffer fault pause ("AFC NOT FEEDING", "CLOG DETECTED") during a print, runs `RESUME` by itself after 5 s, so the lane check and the purge check decide; a failed check keeps it paused. At most 2 per print and none within 10 min of printing after the previous one; not during `PRINT_START` or a PA test. Off: `enabled` in `_AUTO_RESUME_CFG` | `print/pause_resume.cfg`, `purge_check.cfg`, AFC |
| `auto_resume.cfg` | After an AFC "NOT FEEDING" buffer fault pause during a print (not "CLOG DETECTED": usually a real blockage), runs `RESUME` by itself after 5 s, so the lane check and the purge check decide; a failed check keeps it paused. At most 2 per print and none within 10 min of printing after the previous one; not during `PRINT_START` or a PA test. Off: `enabled` in `_AUTO_RESUME_CFG` | `print/pause_resume.cfg`, `purge_check.cfg`, AFC |
| `purge_until_empty.cfg` | `PURGE_UNTIL_EMPTY`: purges a snapped filament out until there is no more nozzle pressure | same as `purge_check.cfg` |
| **pressure_advance/** | | |
| `bd_pressure.cfg` | `[bdpressure bd_pa]` (I2C on the Nitehawk) and the vendor macros `PA_CALIBRATE`, `PA_E`, `PA_RESET`, `PA_STATE`; `PA_CALIBRATE` no longer sets Z back to a value read before the test (wrong after a Z hop in `G28 XY`) | `bdpressure.py` from the bd_pressure repo |
+12 -5
View File
@@ -1,6 +1,6 @@
# Automatic resume after an AFC buffer fault ("AFC NOT FEEDING" / "CLOG DETECTED")
# Automatic resume after an AFC "NOT FEEDING" buffer fault
#
# When AFC pauses a print because the buffer stayed empty or full too long,
# When AFC pauses a print because the buffer stayed empty too long,
# PAUSE (print/pause_resume.cfg) calls _AUTO_RESUME_ARM. After `delay` seconds
# _AUTO_RESUME runs the normal RESUME: AFC clears its error, the lane check
# runs, and _RESUME_PRINT purges into the Blobifier with the purge check. A
@@ -24,7 +24,12 @@ variable_enabled: True
variable_max_per_print: 2 # automatic resumes per print
variable_min_interval: 10 # minutes of printing since the previous automatic resume
variable_delay: 5 # seconds after the pause (AFC finishes its own pause first)
variable_faults: ['NOT FEEDING', 'CLOG DETECTED'] # text in AFC's error message
variable_faults: ['NOT FEEDING'] # text in AFC's error message
# Not 'CLOG DETECTED' (buffer stayed full: the extruder takes no filament):
# that is usually a real blockage. On 2026-11-17 a nozzle blocked right after a
# swap; the purge check still passed (a blocked nozzle also builds pressure),
# so the automatic resume only printed air until the next CLOG pause.
variable_clear_faults: ['NOT FEEDING', 'CLOG DETECTED'] # removed from AFC's queue after any resume
# State (set by the macros)
variable_active: False # True from _PRINT_START_FINISH to the end of the print
variable_count: 0 # automatic resumes in this print
@@ -88,12 +93,14 @@ gcode:
{% endif %}
# Called by _RESUME_PRINT after a resume: remove a buffer fault message from
# the front of AFC's queue, so the next pause reads its own message
# the front of AFC's queue, so the next pause reads its own message. Also a
# CLOG message after a manual RESUME: left at the front, it would hide the next
# NOT FEEDING message and block that automatic resume.
[gcode_macro _AUTO_RESUME_CLEAR_FAULT]
description: Helper: drop AFC's buffer fault message after a resume
gcode:
{% set cfg = printer["gcode_macro _AUTO_RESUME_CFG"] %}
{% set msg = printer.AFC.message.message|default('') if 'AFC' in printer else '' %}
{% if msg and cfg.faults|select('in', msg)|list %}
{% if msg and cfg.clear_faults|select('in', msg)|list %}
AFC_CLEAR_MESSAGE
{% endif %}
+13 -1
View File
@@ -9,6 +9,10 @@
# the nozzle pressure stayed above the threshold. Catches filament
# that never reached the melt zone. Skipped automatically on stock
# firmware or when bdpressure_check.py is not installed.
# A blocked nozzle builds pressure too, so that alone passes it. The
# pressure left when the check stops (delta, a few seconds after the
# purge, once the blob is dropped) catches some: it bleeds off slowly
# through a blocked nozzle (pressure_max_delta).
#
# Needs: bdwidth_status.py in ~/klipper/klippy/extras/
# bdpressure_check.py in ~/klipper/klippy/extras/ (optional)
@@ -46,6 +50,11 @@ variable_motion_check: False # False: skip the bdwidth motion check, pressure
variable_min_ratio: 0.4 # placeholder: sensed mm / extruded mm below this = not feeding
variable_min_extrude: 20 # skip the checks for purges shorter than this (mm)
variable_pressure_min_ms: 5000 # placeholder: longest continuous pressure needed (ms)
# Pressure still above the baseline when the check stops = nozzle (partly)
# blocked; 0 = off. Data 2026-11-12..17: 16 good purges -3..+10, the two long
# purges into a blocked nozzle (2026-11-17) +19 and +21; the two 40 mm purges
# into that nozzle gave -2 and -12, so it does not catch every blockage.
variable_pressure_max_delta: 15
# bdwidth is not reading after SET_BDWIDTH ... COMMAND=DISABLE and after
# every Klipper restart until an ENABLE (outside a print, often early in a
# print start). The check then switches it on and waits this long for the
@@ -188,10 +197,13 @@ gcode:
{% endif %}
{% if 'bdpressure_check' in printer and printer.bdpressure_check.available %}
{% set p = printer.bdpressure_check %}
{% set press = "pressure longest %d ms, total %d ms, peak %d" % (p.longest_ms, p.above_ms, p.peak) %}
{% set press = "pressure longest %d ms, total %d ms, peak %d, left %d" % (p.longest_ms, p.above_ms, p.peak, p.delta) %}
{% if p.longest_ms < cfg.pressure_min_ms %}
{% set _ = failed.append("no sustained nozzle pressure") %}
{% endif %}
{% if cfg.pressure_max_delta > 0 and p.delta > cfg.pressure_max_delta %}
{% set _ = failed.append("pressure stays high after the purge (%d > %d): nozzle blocked?" % (p.delta, cfg.pressure_max_delta)) %}
{% endif %}
{% elif 'bdpressure_check' in printer %}
{% if cfg.ignore_missing_sensor %}
{% set press = "pressure skipped: bd_pressure not answering (ignored, PURGE_CHECK_IGNORE_SENSOR)" %}
+1 -1
View File
@@ -16,7 +16,7 @@
# (the lane AFC was loading when a tool change failed, otherwise the loaded
# lane), and RESUME stops with an error when another lane (or none) is loaded.
# RESUME LANE_CHECK=0 resumes anyway (e.g. after remapping with SET_MAP).
# An AFC buffer fault pause ("AFC NOT FEEDING" / "CLOG DETECTED") during a
# An AFC "NOT FEEDING" buffer fault pause (not "CLOG DETECTED") during a
# print runs this RESUME by itself after a few seconds, within limits
# (afc/auto_resume.cfg).
# A failed bed check in PRINT_START (calibration/bed_scan_check.cfg) pauses with