Skip to content

Workflow Diagrams

Visual flow diagrams showing how SpoolSense handles filament tracking for different tag types and printer configurations.

Single Toolhead

Tags That Track Filament Usage

Supported tags: OpenPrintTag, OpenTag3D

These tags have writable weight fields. After each print, the UPDATE_TAG macro (in your PRINT_END) triggers the middleware to calculate filament used from Moonraker's print history. The deduction is sent to the scanner and stored persistently. The next time that spool is scanned, the scanner writes the updated weight to the tag before proceeding with the normal scan flow.

The tag is the source of truth — Spoolman is updated from the tag data on each scan.

Key points:

  • UPDATE_TAG fires automatically at the end of each print
  • Deductions accumulate on the scanner if multiple prints occur before the next scan
  • The scanner writes the updated weight to the tag on the next scan, then clears the deduction OpenPrintTag / OpenTag3D workflow

Tags That Do Not Track Filament Usage

Supported tags: UID-only (plain NTAG), TigerTag, OpenSpool

These tags have no writable weight field. Filament tracking is handled entirely by Moonraker's built-in Spoolman integration, which syncs usage in real-time during each print.

UPDATE_TAG still fires in PRINT_END but the middleware detects the tag type and skips — no action is taken. Spoolman already has the correct weight from Moonraker's real-time sync.

Key points:

  • Requires [spoolman] configured in moonraker.conf for filament tracking
  • SpoolSense only handles the initial scan (telling Moonraker which spool is active)
  • Moonraker and Spoolman handle all filament usage tracking automatically
  • UPDATE_TAG is a no-op for these tag types — no errors, just a clean skip

UID-Only / TigerTag / OpenSpool workflow

Multi-Toolhead (Toolchanger)

Tags That Track Filament Usage

Supported tags: OpenPrintTag, OpenTag3D

Same concept as single toolhead, but the middleware reads the per-tool filament_weights array from the last completed job. Each tool's usage is sent as a separate deduction to the scanner, keyed by that tool's active spool UID.

Key points:

  • UPDATE_TAG fires once in PRINT_END — handles all tools in one pass
  • Moonraker provides filament_weights per tool index (T0, T1, T2, T3)
  • Middleware maps tool index to active spool UID and sends one deduction per spool
  • Scanner accumulates deductions per UID independently
  • Tools with zero usage in a job are skipped

Multi-Toolhead workflow

Tags That Do Not Track Filament Usage

Supported tags: UID-only (plain NTAG), TigerTag, OpenSpool

The flow is identical to the single toolhead case — UPDATE_TAG is a no-op and Moonraker handles filament tracking for all tools via its built-in Spoolman integration. See the Single Toolhead — Tags That Do Not Track Filament Usage section above.

AFC (Box Turtle / Armored Turtle)

Tags That Track Filament Usage

Supported tags: OpenPrintTag, OpenTag3D

AFC already tracks per-lane remaining weight in real-time using extruder position monitoring. When UPDATE_TAG fires in PRINT_END, the middleware reads the current weight per lane from AFC's status endpoint — no print history calculation needed. The difference between the initial tag weight and AFC's current weight is the deduction.

Key points:

  • AFC tracks weight per lane in real-time (extruder position → mm → grams using filament density)
  • UPDATE_TAG reads AFC weights, not Moonraker print history — more accurate than slicer estimates
  • Middleware calculates deduction per lane: initial tag weight minus AFC current weight
  • One deduction sent per active spool UID, same scanner persistence as other setups
  • Handles multi-lane prints where AFC swaps lanes mid-print

AFC workflow

Tags That Do Not Track Filament Usage

Supported tags: UID-only (plain NTAG), TigerTag, OpenSpool

Same as single toolhead — UPDATE_TAG is a no-op. AFC handles Spoolman tracking via its own spool_id integration per lane. See the Single Toolhead — Tags That Do Not Track Filament Usage section above.

Bambu Lab AMS

Bambu Lab support uses Home Assistant blueprints — no middleware and no UPDATE_TAG macro involved. Two blueprints handle the full flow:

  1. SpoolSense → Bambu AMS — Scan a spool, load it into an AMS tray within 5 minutes, and the blueprint auto-pushes material, color, and temperature data to that tray via bambu_lab.set_filament. Also stores the spool UID and Spoolman ID in HA input helpers for deduction tracking.

  2. SpoolSense → Spoolman Deduction — When a print finishes or is canceled, reads the per-tray weight breakdown from the Bambu print weight sensor, looks up the spool UID per tray from the stored helpers, and calls spoolman.use_spool_filament to deduct usage.

Key points:

  • No middleware, no Moonraker, no Klipper macros — entirely HA-driven
  • Per-tray weight deduction from Bambu's own print weight sensor
  • Requires: HA Bambu Lab integration, spoolman-homeassistant integration, SpoolSense MQTT sensor
  • Supports up to 16 trays across 4 AMS units
  • Physical NFC tag weight is not updated — deduction goes to Spoolman only

Bambu Lab AMS workflow