How to Calibrate a 3D Printer: Flow, Temperature and Steps

Calibration means measuring what your printer actually does and then correcting the settings to match. Done in the right order it removes most causes of rough surfaces, stringing and weak layers. The order matters: temperature changes how much plastic flows, and flow changes how much pressure builds in the nozzle. This guide follows the order recommended in the OrcaSlicer calibration guide and the Klipper documentation. The values given are starting points taken from that documentation, not results from our own printing, and your printer maker's guidance comes first.

Many calibration tests are built into OrcaSlicer, and the same ideas apply in other slicers and firmwares. OrcaSlicer's own pages warn against selecting its flow calibration option on Bambu Lab printers, so check your printer's documentation before running any test.

Before you calibrate: the basics

Calibration cannot fix a loose belt or a dirty nozzle. Check these first:

• Belts and screws are tight and the frame does not rock.
• The nozzle is clean, with no partial clog. See how to fix a clogged nozzle.
• The filament is dry. Wet filament gives misleading results on every test. See how to dry filament.
• The bed is level and the first layer is right. See how to fix a bad first layer and how to level a 3D printer bed.

The calibration order

1. Check extruder steps (esteps / rotation distance)

If the extruder feeds the wrong length of filament, every later calibration is compensating for a hardware error. Klipper documents a "measure and trim" check:

1. Heat the hotend to a suitable temperature with filament loaded.
2. Mark the filament about 70 mm from the extruder intake and measure that distance precisely with calipers (the initial distance).
3. Extrude 50 mm slowly (Klipper's example uses G91 followed by G1 E50 F60). A fast extrusion rate can skew the result.
4. Measure the distance again. Actual extrusion = initial distance minus new distance.
5. New rotation distance = old value x actual / requested (50 mm).

Klipper advises repeating the check if the actual and requested lengths differ by more than about 2 mm, and says not to use this method on X, Y or Z axes. Firmwares that use esteps instead of rotation distance follow the same logic; consult your firmware's documentation.

2. Temperature tower

OrcaSlicer's temperature tower prints blocks at stepped temperatures. Nozzle temperature affects viscosity: low temperature can cause under-extrusion, poor layer adhesion and stringing, and high temperature can cause filament degradation, clogging and stringing. Look for the block with the best balance of stringing, layer adhesion, overhangs and bridging. If a range looks good, use the middle of it, or the higher end if you plan to print faster. Stay inside the temperature range printed on the spool. For stringing in particular, see fix stringing.

3. Max volumetric speed

Max volumetric speed (mm³/s) varies with the material, hotend, nozzle and even filament brand and color, so it needs to be set per filament. OrcaSlicer's default test runs from 5 to 20 mm³/s in 0.5 steps. Print it, find the height where quality fails, and use start + (height measured x step). In the wiki's example, a height of 19 mm gives 5 + (19 x 0.5) = 14.5 mm³/s. The wiki notes this is a best-case result and suggests reducing the setting by 10% to 20% or more for quality and strength. Raising the temperature toward the higher end of the filament's range can help at higher flow, as the temperature tower shows.

4. Pressure advance

Pressure advance compensates for the lag in extrusion pressure when the print head changes speed. OrcaSlicer offers tower, pattern and line methods, with separate versions for direct-drive and Bowden extruders. For the tower method, the value rises with height from the start value in set steps: value = start + (step x measured height). The wiki example gives 0 + (0.002 x 8) = 0.016. Marlin printers need Linear Advance enabled in firmware, and Klipper has pressure advance built in.

5. Flow ratio

Flow ratio controls how much filament is extruded. Too low gives under-extrusion, gaps and weak layers. Too high gives excess material and rough surfaces. OrcaSlicer's recommended method prints eleven blocks with different modifiers (a range of -0.05 to +0.05 in 0.01 steps for the recommended setting). Pick the block with the smoothest top surface and no gaps between the arcs, then update: new flow ratio = old flow ratio ± modifier. In the wiki's example, an old value of 0.98 and a best block of +0.01 gives 0.99. Save the filament profile afterward.

6. Retraction

Retraction tower tests print a vertical structure with notches at different retraction lengths. The best length is the shortest that gives the cleanest tower. OrcaSlicer's defaults run from 0 to 2 mm in 0.1 mm steps, suited to most direct-drive extruders; for Bowden setups the wiki suggests 1 to 6 mm in 0.2 mm steps. Do this after flow and pressure advance, as the Orca guide recommends.

7. Optional advanced steps

The guide also lists cornering (jerk and junction deviation), input shaping to reduce ringing, a VFA speed test and a tolerance calibration so parts fit together accurately. These matter most when you push speed. If you see ripples after corners, read how to fix ghosting and ringing.

A quick test to confirm it worked

Print a small test cube or calibration model and measure it with calipers, and look at the top surface and corners. Compare it with a print from before. If dimensions are consistent and top layers are smooth, save the profile and name it clearly (brand, color, nozzle). Repeat only the temperature, flow and pressure advance steps when you change filament. If prints still fail, see common 3D printer problems.