You start flow calibration from the default value of 100%, and the target number for your printer and filament almost always falls somewhere between 90% and 110%. It's one of the most important steps when setting up a new spool, especially when switching to a different material, for example from PLA to a technical filament — and one of the most often skipped.
This guide walks you through printing a simple test model, measuring it with a caliper, and calculating a new flow value step by step — no guesswork, and no expensive trial and error on large models.
Starting settings for flow calibration — table
| Parameter | Value | Comment |
|---|---|---|
| Layer height | 0.2 mm | Standard value for the calibration test |
| Number of walls (perimeters) | 2 | Enough thickness for a reliable measurement |
| Infill | 0% | The test only concerns the outer walls |
| Top layers | 0 | The model is left open at the top, not closed |
| Bottom layer | 1 | A stable base for the test print |
| Test model dimensions | 30×30×20 mm | A simple cube with no details or corners that complicate measurement |
| Starting flow rate | 100% | The default value in most slicers |
| Typical correction range | 90–110% | You rarely need to go beyond this range |
| Outer wall print speed | 30–40 mm/s | Slower printing gives a more accurate, repeatable measurement |
| Nozzle temperature | depends on the material | See the temperature table for full values |
| Measuring tool | digital caliper (0.01 mm) | A cheap analog caliper usually isn't accurate enough |
| Measurements per wall | minimum 3 | Average the result to eliminate a random error |
You'll find full nozzle and bed temperature values for individual materials in the print temperature table — it's worth keeping it open next to your slicer during the whole calibration process.
Why flow calibration matters so much
Flow (sometimes called the extrusion multiplier) determines how much material the printer actually extrudes relative to what the slicer "assumes" based on the programmed filament diameter and nozzle geometry. In theory, a value of 100% should be ideal for every printer. In practice, the result is affected by dozens of small factors: the actual filament diameter (rarely a perfect 1.75 mm along its entire length), nozzle wear, the pressure of the extruder's drive gear, or e-steps calibration.
An incorrectly set flow shows up first exactly where you expect precision: on the first layer, on the outer walls, and on the top surfaces of a model. Too much flow means excess material — bulging, shiny walls, blurred details, sometimes a slight increase in dimensions. Too little flow means under-extrusion — visible gaps between print lines, weak layer bonding, and porous top surfaces, even if the rest of your settings are correct.
This is exactly why it's worth calibrating flow right after setting up your first layer — the two calibrations complement each other, and only together do they give you a stable base for further tuning of the printer, such as speed or retraction.
What flow actually is and what it affects
Flow vs. extrusion multiplier — are they the same thing?
In most popular slicers (PrusaSlicer, Bambu Studio, OrcaSlicer, Cura), "flow" and "extrusion multiplier" are, in practice, the same value, just named differently and sometimes expressed differently: as a percentage (100%) or as a multiplier (1.00). Changing one of these values works exactly the same way — it multiplies the volume of material extruded per millimeter of nozzle movement.
How flow affects the dimensions and look of a print
Flow doesn't directly change speed or temperature — it only affects the amount of material. That's why a well-calibrated flow primarily improves the dimensional accuracy of walls, the quality of top layers, and the strength of bonds between print lines. If a model needs to fit another part with a tolerance of a few tenths of a millimeter, flow calibration is one of the first steps worth starting with.
The calibration test step by step
Preparing the test model
The simplest model is a hollow cube measuring 30×30×20 mm, printed with two walls, no infill, and no top layers (settings from the table above). An alternative is Vase mode (a single-wall spiral print) on a 40×40×40 mm model — it needs less material and time, but then you're measuring a single wall instead of a double one.
Printing and measuring with a caliper
After printing, wait for the model to cool to room temperature — measuring a warm part can be distorted by thermal expansion. Measure the wall thickness with a caliper in at least three places at the middle height of the model (avoid the very bottom and very top, where start and end-of-print inaccuracies can occur). Record all the measurements and calculate the average.
Calculating the new flow value
The formula is simple: new flow = (expected wall thickness ÷ measured wall thickness) × current flow. Expected wall thickness is usually the extrusion width set in the slicer (by default close to the nozzle diameter, e.g. 0.42–0.45 mm for a 0.4 mm nozzle) multiplied by the number of walls. If the measured value is smaller than expected, the new flow will come out higher than 100% — and vice versa.
Verifying and locking in the result
After entering the new flow value in your slicer, don't treat it as final right away. Print the test model again with the corrected value and measure it exactly the same way as the first time — the same three measurement points on the wall, the same cooling time before measuring. If the second attempt comes very close to the expected thickness (a difference of around 0.01–0.02 mm), you can consider the result final. A bigger discrepancy usually means it's worth repeating the calculation based on the new measurement instead of continuing to adjust the value "by eye".
Save the established flow value in the material profile in your slicer, not just in the current project — that way you don't start calibration from scratch the next time you print with the same filament. If you use several printers of the same model, it's also worth noting the result separately for each one, since even with an identical hardware configuration, small differences between individual units are normal.
The visual method — calibration without a caliper
If you don't have an accurate caliper on hand, you can calibrate flow using the visual method. Print the test model with a closed top (a few top layers) and examine its surface at an angle to the light. Visible gaps and see-through spots between the top layer's infill lines mean under-extrusion — increase flow by 1–2% and repeat the test. Shiny, bulging streaks of material and small "whiskers" along the edges mean over-extrusion — decrease flow by 1–2%. The visual method is less precise than measuring with a caliper, but it's more than enough for everyday printing of decorative and functional models without high dimensional requirements.
Flow calibration for different materials
PLA and PLA+
PLA is usually the most predictable material to calibrate — its stable viscosity across the typical printing temperature range makes for repeatable measurements. It's a good starting point if you're just learning the procedure, before you move on to more demanding materials, for example high-impact PLA.
PETG and HT-PETG
PETG has a higher melt viscosity than PLA, so it tends to be more sensitive to small temperature changes during the test — it's worth sticking to a single, fixed nozzle temperature from the temperature table throughout the entire calibration process, so the measurement isn't distorted by changing material viscosity.
ABS, ASA, and technical materials
Materials printed in an enclosed chamber need extra attention: the ambient temperature inside the enclosure affects cooling and the behavior of the molten material as it leaves the nozzle. If you change the chamber temperature between prints, it's worth repeating flow calibration after the new conditions have stabilized, rather than assuming the result from before the change is still valid.
Settings for popular configurations
| Configuration | Suggested starting point | Comment |
|---|---|---|
| Bowden | 100%, adjust in 2–3% steps | A longer filament path increases the risk of momentary under-extrusion during fast moves |
| Direct drive | 100%, usually smaller adjustments | A shorter filament path gives a more stable, repeatable flow |
| CoreXY with an enclosed chamber | calibrate after the chamber heats up | Higher ambient temperature changes the material's viscosity at the nozzle outlet |
| Multi-spool system (e.g. AMS/CFS) | calibrate each slot separately | Individual spools of the same material can differ slightly |
| Printers with a filament flow sensor | don't skip flow calibration | The sensor monitors continuous feeding, but doesn't correct your slicer settings |
Most common mistakes
Calibrating on a single spool and assuming the result is universal. Record the flow value separately for each color and production batch of material, instead of assuming one value for an entire material.
Measuring a print that hasn't cooled down yet. Wait until the model reaches room temperature before reaching for the caliper.
Making too large a correction at once. Change flow in steps of 1–3%, not by a dozen percent at once, and repeat the full test each time.
Confusing the flow setting with an independent flow setting for a specific print element. Make sure you're correcting the global material value, not a local setting visible only in one profile.
Calibrating flow with uncalibrated extruder e-steps. If e-steps are badly off, flow calibration only masks the problem instead of solving it — it's worth checking basic extruder calibration first.
Ignoring nozzle temperature during the test. Stick to a single, recorded temperature throughout the process — changing it between test prints skews the result.
Measuring at a single point on the wall instead of several. A single measurement easily lands on a local irregularity — always average at least three readings.
Frequently asked questions
Do I need to calibrate flow for every spool separately?
In an ideal world, yes — different production batches of the same material can differ slightly in diameter and density. In practice, it's enough to calibrate once for a given material and color, and repeat a quick visual test if you notice a change in print quality after loading a new spool.
How often should I repeat flow calibration?
It's worth returning to the test after any major change to the printer: a nozzle or extruder replacement, e-steps calibration, or switching to a completely different material. With a stable configuration and the same material, the value usually stays valid for a long time.
Can bad flow cause warping or a poor first layer?
Indirectly, yes — too much material on the first layer makes it harder for it to adhere evenly to the bed, while too little weakens its mechanical strength. Flow alone is rarely the sole cause of warping, but a poorly calibrated one definitely makes it harder to diagnose other problems.
Does flow calibration replace extruder e-steps calibration?
No. E-steps determine how much material the extruder motor physically feeds per revolution, while flow is an additional multiplier in the slicer. The correct order is to calibrate e-steps first, and only then fine-tune flow.
What should I do if I still see gaps between walls after calibration?
Check whether the problem lies elsewhere: nozzle temperature too low, outer wall print speed too high, or layer height too large relative to the nozzle diameter. Flow corrects the amount of material, but it won't replace correct temperature and speed settings.
Summary
Flow calibration is one of those tasks you do once per material and benefit from for a long time. A simple test model, a caliper, and one simple formula are enough to rule out under- and over-extrusion as the cause of problems with a print's dimensions and appearance.
If you're just starting a test on a new material, it's worth doing it on a proven, stable filament from the PLA filament collection — that way you can be sure the calibration result reflects your printer's settings, not instability in the material itself.