Calibrate a filament 3D printer in this order: check the mechanics, run the machine’s bed and first-layer routine, tune the filament profile, then measure a test part. Change one setting at a time and save the result. A calibration cube is a final check, not a reason to change every motor setting.
This guide covers FDM/FFF filament printers. Resin printers need a separate workflow involving build-plate setup and resin exposure. The procedure below is a synthesis of manufacturer and firmware documentation, checked September 11, 2026; it synthesizes established calibration procedures rather than individual bench tests of every printer.
Before changing any settings
Save a copy of the working printer and filament profiles. Record the printer, firmware, nozzle diameter, build plate, filament and slicer version. Start from the correct manufacturer’s profile rather than a profile for a vaguely similar machine.
With the printer cool and powered down where its manual requires, check for loose pulleys, damaged belts, play in the toolhead, a badly seated plate and a clogged or worn nozzle. Check that the spool feeds freely. Tighten or adjust only to the manufacturer’s instructions: excess belt tension creates another fault.
A print that suddenly shifts sideways needs a mechanical and motion check. Use our four-pattern layer-shift diagnosis before trying to correct it with extrusion settings.
1. Level the bed and establish the first layer
Clean and install the build surface using its approved care method. Run the printer’s supplied calibration sequence at the temperatures it specifies. If the bed has adjustment knobs, tram it as instructed before creating a bed mesh. A mesh compensates for measured surface variation; it cannot repair a loose bed.
Do not assume every automatic-leveling printer needs a manual Z offset. Prusa’s Live Adjust Z documentation distinguishes older machines from load-cell models such as the MK4 and XL that set the first layer automatically. Use the routine for your exact printer.
On a machine requiring manual offset, print its supplied first-layer pattern. Separated strands or weak attachment can indicate too much distance; ridges, scraping or an excessively thin layer can indicate too little. Stop if the nozzle contacts the plate. Make small adjustments through the documented control, then repeat the same pattern.
For a Klipper probe setup, follow Probe Calibrate , including its distinction between probe offset and bed mesh. Do not paste another owner’s offset into your configuration.
2. Check extrusion only when there is evidence of a feed error
Extruder feed calibration and slicer flow ratio solve different problems. A feed setting describes how motor movement advances filament. Flow ratio adjusts how much material the slicer requests for a particular filament profile.
On a supported open configuration, a measured filament-feed test can identify a wrong extruder setting. Follow the firmware’s procedure, including extrusion temperature and feed rate. Do not change locked or factory-managed motor settings on an appliance-style printer just because a wall measurement differs from an online example.
Klipper’s rotation-distance guide explains the extruder measurement procedure. It also cautions against measuring printed objects to recalibrate X, Y and Z rotation distance: those axes normally derive from the machine’s belt, pulley or screw geometry.
3. Tune the filament profile in a controlled sequence
Use one dry, known spool and the correct nozzle profile. Keep the bed surface, cooling and test geometry consistent. If wet filament is popping or producing an unusually rough surface, address that before interpreting a calibration pattern.
| Setting | What to inspect | What it cannot fix |
|---|---|---|
| Nozzle temperature | Layer bonding, surface quality and bridging across a suitable test range | Loose belts or a dirty plate |
| Flow ratio | Gaps or excessive ridges in the slicer’s flow test | An incorrect first-layer offset |
| Pressure advance | Corner and line-width transitions in the supported test | An obstructed filament path |
| Retraction | Travel strings after temperature and flow are stable | Every moisture or nozzle problem |
| Speed and volumetric flow | Loss of quality as requested output rises | A promise that headline travel speed is usable print speed |
Use the tests and interpretation instructions in your slicer, such as the official OrcaSlicer calibration guide . Keep the best result from one test before moving to the next. There is no single temperature, flow ratio or retraction distance that is correct for all PLA and all extruders.
4. Measure a test part without corrupting axis geometry
Print a modest test object with known outside dimensions and an example of the hole or mating feature that matters to your project. Let it cool before measuring. Record the requested dimension, measured dimension, material and orientation. Repeat a measurement rather than trusting one caliper placement.
If the bottom edge is wider than the upper walls, investigate first-layer compression and elephant-foot compensation. If outside dimensions are close but holes are tight, investigate the slicer’s hole or horizontal compensation and the design’s intended clearance. Material shrinkage and extrusion width also affect fit.
Do not adjust X/Y/Z motor steps from a single cube. A geometry change made to compensate for one material can make another part worse. Establish mechanical correctness first, then use material or feature compensation where the evidence supports it.
Save a calibration record
Keep a short note with five fields: symptom, test, setting before, setting after, result. For example, record whether a first-layer pattern changed from separated lines to a joined surface after the documented offset adjustment. This is a recording method, not a test result from our lab.
Retest the relevant stage after a nozzle, build surface, extruder or firmware change. A new filament usually calls for a filament-profile check, not a complete rebuild of the printer configuration.
Frequently Asked Questions
What should I calibrate first?
Check the mechanics and installed profile, then follow the printer’s bed and first-layer routine. Flow tuning is hard to interpret while the nozzle-to-bed relationship is wrong.
Does automatic bed leveling calibrate everything?
No. It measures or compensates for the bed relationship. Filament temperature, flow, pressure advance and dimensional fit remain separate checks, although some printers automate additional stages.
How accurate should a calibration cube be?
There is no universal pass tolerance for every printer and application. Compare repeated measurements with the clearance your actual part needs, and investigate the pattern of error before changing settings.