From 52c8966c8f1d103da38da89cbb7d595888dad288 Mon Sep 17 00:00:00 2001 From: jessikitty Date: Tue, 28 Jul 2026 11:39:51 +1000 Subject: [PATCH] Document self-test usage, safety and stages --- README.md | 48 ++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 46 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index c97d006..01570ed 100644 --- a/README.md +++ b/README.md @@ -1,3 +1,47 @@ -# cr-x-mechanical-selftest +# CR-X Mechanical / Thermal / Extrusion Self-Test -Non-printing mechanical, thermal and extrusion self-test G-code for the Creality CR-X (run via OctoPrint before calibration) \ No newline at end of file +Non-printing G-code to exercise every subsystem of a **Creality CR-X** before +running OrcaSlicer calibrations. Run via **OctoPrint** (upload, then print the +file — nothing is deposited on the bed). + +## Files +| File | What it does | Heating? | Extrudes? | +|------|--------------|----------|-----------| +| `CR-X_selftest_cold_motion_only.gcode` | Homing, X/Y/Z travel, diagonals, corners, speed range | No | No | +| `CR-X_selftest_full.gcode` | Everything above **plus** bed heat, hotend heat, fan ramp, extrude/retract | Yes | Yes | + +**Run the cold one first.** If motion is clean and quiet, run the full test. + +## Before you start +- **Stay with the printer** the whole time. Keep the power switch / e-stop in reach. +- For the **full** test, **load filament** first (the extrude stage feeds ~90 mm total). ABS or any filament is fine — nothing is printed. +- Clear the bed and gantry of tools/clips. + +## Assumptions (edit if yours differ) +- Build volume **300 × 300 × 400 mm**. All moves keep a **5 mm margin** from each axis limit and **20 mm** off Z-max. +- Bed heated to **100 °C** (CR-X rated ≤100 °C). +- Hotend heated to **230 °C** only — safe for the stock **PTFE-lined** hotend and hot enough to extrude ABS without grinding. Cold extrusion is never commanded. +- Bowden retraction test uses **6 mm**. + +If your CR-X has a filament runout sensor that halts on the extrude test, disable it for the run or ignore the trigger. (The `M runout ignore` line in the full file is a comment placeholder, not a command — remove it if your firmware flags unknown lines; most Marlin builds just skip it.) + +## What each stage checks (watch the LCD `M117` messages) +1. **Homing** — endstops seek/back-off/re-tap. Listen for grinding or a crash. +2. **X / Y sweeps** — full-width and full-depth travel, belts smooth. +3. **Z sweep** — **both lead screws must turn together**; gantry stays level. +4. **Diagonals + corners** — combined XY motion, no binding. +5. **Speed check** — fast then slow moves; listen for chatter/resonance. +6. **Bed heat 100 °C** — smooth climb on the OctoPrint graph = good thermistor/heater. +7. **Hotend heat 230 °C** — same, no wild swings (that would mean a PID/thermistor issue). +8. **Fan ramp 50 % → 100 % → off** — part-cooling fan spins clean, no bearing rattle. +9. **Extrude/retract** — steady flow, no clicking or skipping on the Bowden gears; the 6 mm retract should visibly pull back. +10. **Cool down + park** — heaters off, gantry dropped, bed presented, steppers disabled. + +## If something fails +Note the **`M117` stage name** shown on the display when it went wrong — that pinpoints the subsystem. Common ones: +- Axis grinds at home → endstop or belt/pulley. +- One Z screw lags → coupler grub screw or binding rod. +- Temp graph swings or bails with a MINTEMP/MAXTEMP → thermistor/heater wiring. +- Extruder clicks → nozzle not hot enough, clog, or filament not loaded. + +After a clean pass, move on to PID tuning and the OrcaSlicer calibration workflow.