Amalgam Calibration Cheat Sheet

Quick reference for Klipper calibration commands. For detailed explanations, see calibration-guide.md.

The 80/20 Path (Quick Start)

Just want to print? Do these 3 things and you’re 80% calibrated:

Step Print Time What to Check
1. First Layer See options below 2-10 min Lines touch, smooth surface, consistent
2. Dimensions xyz_calibration_cube.stl 15 min Measures 20.0mm on all axes
3. Quality 3DBenchy.stl 60 min No stringing, good overhangs, clean detail

First layer options: - first_layer_patch_0.2mm.stl - Quick center check (~2 min) - first_layer_grid.stl - Full bed test (~10 min) - prusa_live_z.stl - Adjust Z live while printing (~8 min)

If all three pass: You’re ready to print. Do the advanced tuning later (or never).

If something’s wrong: Check the troubleshooting table at the bottom, then do the relevant advanced calibration.

# Generate parametric calibration prints (uses your bed size)
./build.sh build first_layer_grid   # Grid of squares
./build.sh build prusa_live_z       # Prusa-style live Z pattern

# Download standard calibration prints
python utilities/download_calibration.py

Simple vs Advanced Path

Path Prints Time Result
Simple (80/20) First layer → XYZ cube → Benchy ~90 min Good prints, works for most people
Advanced All 10 steps below 3-4 hours Optimized PA, Input Shaper, perfect corners

When to go Advanced: - Printing functional parts with tight tolerances - Seeing ringing/ghosting artifacts - Corners are bulging or have gaps - Want maximum print speed

Parametric Calibration Prints (build123d)

Generated to match your bed size from config.py:

Model Command Use For
first_layer_grid.stl ./build.sh build first_layer_grid Grid of squares across bed
first_layer_snake.stl (generated with grid) Alternative snake pattern
prusa_live_z.stl ./build.sh build prusa_live_z Live Z adjustment while printing

First Layer Test Options

Pattern Source Best For
first_layer_patch_0.2mm.stl Ellis (download) Quick center check (~2 min)
first_layer_grid.stl Amalgam (build123d) Full bed coverage, tests mesh leveling
prusa_live_z.stl Amalgam (build123d) Live Z adjustment while printing

Quick check? Use Ellis patch (center only). Testing bed mesh? Use grid (covers entire bed). Tuning Z-offset? Use Prusa pattern (adjust while printing).

Downloaded Calibration Models

Run python utilities/download_calibration.py to download these:

Model Size Use For Source
first_layer_patch_0.2mm.stl 0.2mm First layer / bed level Ellis
extrusion_multiplier_cube.stl ~25mm Flow rate calibration Ellis
xyz_calibration_cube.stl 20mm Quick dimensional check 5axes
voron_design_cube.stl 30mm Advanced calibration Voron
square_tower.stl ~80mm PA, Temperature Klipper
ringing_tower.stl ~80mm Input Shaper Klipper
3DBenchy.stl ~48mm Final validation CreativeTools

Key: TUNING_TOWER requires a tall object. Use square_tower.stl, NOT the cubes.

XYZ Cube vs Voron Cube

Feature XYZ Cube (20mm) Voron Cube (30mm)
Dimensional check Yes Yes
Overhang test No Yes (logos)
Bridging test No Yes (internal)
Bearing fit test No Yes (625, F695)
PA corner check No Yes (90° corners)
Print time ~15 min ~30 min

Use XYZ cube for quick dimensional checks. Use Voron cube when tuning for functional parts with bearing fits and overhangs.

Voron Cube vs Benchy

Test Voron Cube Benchy
Dimensions Yes (30mm) Yes (60mm hull)
Overhangs Yes (logos) Yes (hull, multiple angles)
Bridging Yes (internal) Yes (cabin roof)
Fine detail No Yes (text, windows)
Curved surfaces No Yes (hull)
Stringing test No Yes (chimney)
Bearing fits Yes No
PA corners Yes No
Universal reference No Yes (compare to millions online)

Voron cube = “Can I print functional parts?” (tolerances, bearing fits) Benchy = “Does my print look right?” (visual reference everyone knows)

Recommendation: Print Voron cube for functional validation. Print Benchy if you want visual comparison or suspect quality issues. Print Fidget Bolt as the real victory test (working threads = well calibrated).

Calibration Order (Advanced Path)

# Step Path Required?
1 Mechanical check Both Yes
2 PID tune hotend & bed Both Yes (once)
3 Z-offset & bed mesh Both Yes
4 First layer test Simple Yes
5 Extruder steps (e-steps) Advanced If dimensions off
6 Pressure Advance Advanced If corners bulge
7 Input Shaper Advanced If ringing/ghosting
8 Flow rate Advanced If walls wrong thickness
9 Temperature Advanced Per filament
10 Retraction Advanced If stringing

Simple path: Steps 1-4, then print XYZ cube and Benchy to validate.

1. PID Tuning

Stabilizes temperature control. Run once per hotend/bed change.

# Hotend (at typical print temp)
PID_CALIBRATE HEATER=extruder TARGET=210

# Bed
PID_CALIBRATE HEATER=heater_bed TARGET=60

# Save results
SAVE_CONFIG

2. Z-Offset & Bed Mesh

G28                      # Home all
PROBE_CALIBRATE          # Interactive Z-offset (paper test)
# Adjust with TESTZ Z=-0.1 or TESTZ Z=+0.1
ACCEPT                   # When paper drags slightly
SAVE_CONFIG

# Then create bed mesh
G28
BED_MESH_CALIBRATE
SAVE_CONFIG

Live Z adjust during print:

SET_GCODE_OFFSET Z_ADJUST=-0.02 MOVE=1   # Lower (more squish)
SET_GCODE_OFFSET Z_ADJUST=+0.02 MOVE=1   # Higher (less squish)

3. First Layer Test

Print the Ellis first layer patch (first_layer_patch_0.2mm.stl).

Symptom Fix
Lines don’t touch Lower Z-offset
Rough/transparent Lower Z-offset
Ridges between lines Raise Z-offset
Varies across bed Re-run BED_MESH_CALIBRATE

4. Extruder Steps (E-steps)

Ensures extruder pushes exactly 100mm when asked.

# Heat hotend
M104 S210

# Mark filament 120mm from extruder entrance
# Extrude 100mm slowly
G91                      # Relative mode
G1 E100 F100             # Extrude 100mm at 100mm/min
G90                      # Back to absolute

# Measure remaining distance to mark
# If 20mm remains = perfect. If 18mm remains = over-extruding.

Calculate new e-steps:

new_e_steps = current_e_steps × (100 / actually_extruded)

Update rotation_distance in printer.cfg (inverse relationship).

5. Pressure Advance

Compensates for filament pressure lag. Eliminates bulging corners.

Print: square_tower.stl (~80mm tall)

# Set conservative speed limits
SET_VELOCITY_LIMIT SQUARE_CORNER_VELOCITY=1 ACCEL=500

# Start tuning tower - parameter changes with Z height
TUNING_TOWER COMMAND=SET_PRESSURE_ADVANCE PARAMETER=ADVANCE START=0 FACTOR=.005

Examine the tower walls. Find the height where corners look best (no bulging, no gaps).

PA = START + (height_in_mm × FACTOR)
PA = 0 + (10mm × 0.005) = 0.05

Typical values: | Filament | PA Range | |———-|———-| | PLA | 0.02 - 0.08 | | PETG | 0.05 - 0.10 | | TPU | 0.10 - 0.20 |

Add to printer.cfg:

[extruder]
pressure_advance: 0.05

6. Input Shaper

Eliminates ringing/ghosting. Klipper’s superpower.

Option A: With accelerometer (recommended)

SHAPER_CALIBRATE
SAVE_CONFIG

Option B: Manual with ringing tower

# Print ringing_tower.stl at 80-100mm/s
# Examine for ringing patterns
# Try different shapers:
SET_INPUT_SHAPER SHAPER_FREQ_X=40 SHAPER_FREQ_Y=40 SHAPER_TYPE=mzv

Add to printer.cfg:

[input_shaper]
shaper_freq_x: 40
shaper_freq_y: 40
shaper_type: mzv

7. Flow Rate / Extrusion Multiplier

Print Ellis extrusion_multiplier_cube.stl (single wall, 0 infill, 0 top layers).

Measure wall thickness with calipers. Should equal line width (typically 0.4mm).

new_flow = (expected_width / measured_width) × current_flow
new_flow = (0.4 / 0.44) × 100% = 91%

Adjust in slicer, not Klipper.

8. Temperature Tuning (Optional)

Print: square_tower.stl (~80mm tall)

# Temperature varies with Z height
# START=190, FACTOR=2 means: 190°C at Z=0, 200°C at Z=5mm, 210°C at Z=10mm...
TUNING_TOWER COMMAND=SET_HEATER_TEMPERATURE HEATER=extruder PARAMETER=TARGET START=190 FACTOR=2

Examine each 5-10mm section of the tower: - Too cold: Poor layer adhesion, rough surface, weak - Too hot: Stringing, blobbing, discoloration

Mark the height of the best-looking section, calculate temp: temp = 190 + (height × 2)

9. Retraction Tuning (Optional)

For direct drive (like Pitan), start with: - Distance: 0.5-1.0mm - Speed: 30-45mm/s

Test with stringing torture test or travel-heavy print.

PETG: Less retraction than PLA (more viscous) TPU: Minimal or zero retraction

Validation Prints

After calibration, print these to verify:

Print File What to Check
XYZ Cube xyz_calibration_cube.stl Quick check: 20.0mm on X, Y, Z
Voron Cube voron_design_cube.stl Full check: 30.0mm, overhangs, bearing fits
3DBenchy 3DBenchy.stl Overhangs, bridging, detail, stringing
Maker Coin ./build.sh build maker_coin Logo clarity, layer adhesion
Fidget Bolt ./build.sh build fidget_bolt Threads work = well calibrated!

Quick Troubleshooting

Problem Likely Cause Fix
First layer doesn’t stick Z too high Lower Z-offset
Elephant foot Z too low Raise Z-offset
Ringing/ghosting No Input Shaper Run SHAPER_CALIBRATE
Bulging corners No Pressure Advance Calibrate PA
Stringing Retraction/temp Tune retraction, lower temp
Layer shifts Loose belts, too fast Tighten belts, reduce accel
Inconsistent extrusion Flow rate wrong Calibrate e-steps and flow

External Resources

Download Calibration Prints

cd cad
python utilities/download_calibration.py

Downloads from Klipper, Voron, and Ellis GitHub repos. See calibration-guide.md for print settings.

Credits & Acknowledgments

Amalgam’s calibration system stands on the shoulders of giants:

Source Contribution License
Klipper3d Ringing tower, square tower, TUNING_TOWER GPL-3.0
VoronDesign Voron Design Cube v7 GPL-3.0
Ellis (AndrewEllis93) First layer patch, EM cube, tuning methodology —
5axes XYZ Calibration Cube —
CreativeTools 3DBenchy Public Domain
Prusa Research Live Z pattern concept —
Teaching Tech Calibration methodology —

Amalgam-created prints (first_layer_grid, prusa_live_z) are original build123d implementations inspired by community patterns. Licensed under the same terms as the Amalgam project.

“Measure twice, print once. Or just print it and iterate.”