When a print looks unfinished — thin, gappy walls, barely visible layers, a top surface dotted with tiny holes — the culprit in most cases is under-extrusion, and the first two things to check are a flow rate around 95–105% and a nozzle temperature raised by 5–10 °C above the filament maker's default. Before you start turning sliders in your slicer, though, check whether the problem lies elsewhere — a partially clogged nozzle produces exactly the same symptoms, and tweaking software settings won't fix anything in that case.
This guide walks you through a full under-extrusion diagnosis step by step: a quick test that shows in a few minutes whether the problem is mechanical or software-related, the most common causes — a clogged nozzle, a loose extruder idler, wet filament, bad e-steps calibration — and finally concrete starting values for different printer setups. You'll find a settings table you can copy straight into your slicer, plus a list of mistakes that usually make diagnosis take longer instead of shorter.
Starting settings for under-extrusion diagnosis — table
Before you change anything permanently, go through the table below. It's a set of starting values that work for most home FDM printers — treat them as a starting point for further calibration, not a ready-made recipe.
| Parameter | Value | Comment |
|---|---|---|
| Flow rate | 95–105% | starting point, adjust after a single-wall test print |
| Nozzle temperature — PLA | 205–215 °C | raise by 5–10 °C if layers look translucent and barely bond together |
| Nozzle temperature — PETG | 235–245 °C | same mechanism as PLA, just a higher base range |
| Line width | min. 85% of nozzle diameter | a smaller value deliberately reduces flow — that's not a fault |
| Extruder idler tension | medium, no play | check by hand — the filament shouldn't slide without resistance |
| E-steps test | G1 E100 command | compare the commanded 100 mm against what's actually fed |
| Print speed on thin walls | lower by 20–30% | reduces the volumetric flow required in narrow spots |
| Filament storage | dry box with silica gel | wet filament cuts effective output by 10–30% |
| Volumetric flow limit (0.4 mm nozzle) | approx. 12 mm³/s | above this, a standard hotend usually can't keep up |
| Calibration wall test | 1 perimeter, 0% infill | the basis for the flow calibration test described below |
You'll find full temperature values for each material in our print temperature table — a good reference before you start raising temperature blindly.
Why the printer isn't feeding enough material
Under-extrusion is always a gap between how much filament the slicer commands and how much physically ends up on the bed. The extruder motor turns a set number of steps, but each of those steps can translate into less material than expected — because the gear teeth are slipping, because the nozzle is partially blocked, or because the filament in the hotend hasn't fully melted before it reaches the outlet.
That's why simply "bumping up" flow rate in the slicer rarely fixes the problem for good. If the cause is mechanical — a worn nozzle, a loose idler, wet filament — raising flow rate only masks the symptom, and it comes back at higher print speeds. Proper flow calibration, covered in our step-by-step flow calibration guide, only makes sense once the printer's mechanics are working correctly.
On the other hand, mechanics alone aren't enough if temperature is too low — filament that's too cold is too thick to flow out of the nozzle at the commanded speed, so even a healthy extruder can't deliver as much material as the model needs at a given print speed. That's why it's always worth running diagnosis in a specific order, which we cover in the rest of this guide.
Symptoms of under-extrusion — how to spot it
Under-extrusion rarely shows up in isolation — you usually see several symptoms at once, and it's that combination that helps tell it apart from other problems like ghosting or warping.
- Thin, "lace-like" walls with visible gaps between print lines.
- A top surface with small holes despite correct infill (pillowing).
- Layers bond weakly and the model cracks under light pressure.
- A clicking or ticking sound coming from the extruder during printing.
- Uneven wall thickness — normal in some spots, noticeably thinner in others.
If you recognize at least two of these symptoms at once, you can jump straight to the mechanical-causes section — that's usually where the root of the problem lies.
Mechanical causes of under-extrusion
Mechanical causes are usually the first thing worth checking, because no slicer calibration will fix them. Below are the two most common.
A clogged or partially clogged nozzle
A partial clog doesn't block flow completely — which is exactly what makes it misleading. It shows up mainly at higher print speeds: at slow head movement the nozzle still keeps up, but as speed increases, flow suddenly drops. A brass nozzle printing PLA wears noticeably after roughly 500 hours of use, while abrasive materials — carbon-fiber or wood-filled filaments — can damage a standard nozzle within tens of hours. If you print a lot of technical materials, it's worth considering a hardened nozzle.
A worn nozzle and loose extruder idler
A worn nozzle loses its precise outlet shape, which changes the actual cross-section the molten filament flows through — the effect is similar to a partial clog, only permanent. A separate problem is idler tension: a loose idler lets the filament shift sideways, so instead of pushing material forward, the drive gear slips against its surface. You can check this by looking at the filament segment coming out of the extruder — if it's flattened, chewed, or has visible gouges in one spot, idler tension is the problem. It's also worth checking whether the gear teeth are clogged with filament dust, since that reduces grip just as much as a loose idler.
Material causes of under-extrusion
Sometimes neither the extruder nor the settings are to blame — just the filament itself, how it's stored, or its manufacturing quality.
Wet filament
Hygroscopic filament (PETG, nylon, TPU, but also PLA to a lesser degree) absorbs moisture from the air when stored without an airtight container. In the hotend, that moisture turns to steam and pops as it exits the nozzle — a process known as crackling or hissing filament. Effective material output can then drop by as much as 10–30%, even though the printer itself is working fine. You'll find a full list of symptoms and drying methods in our guide to wet filament.
A second material factor is diameter tolerance — even small deviations from nominal filament diameter along the spool change the actual volume of material fed at the same number of extruder rotations. Filament produced with tight diameter tolerance gives a more predictable, repeatable flow than cheaper alternatives with a wide spread.
Software settings and calibration
It's worth doing calibration only after ruling out the mechanical and material causes described above — otherwise you risk "fixing" with settings a problem that will come back the moment you change filament or speed.
Flow rate calibration
The simplest test is printing a test wall: one perimeter, 0% infill, flow rate set to 100%. After printing, measure the wall thickness in several spots with calipers, away from corners, and compare the average against the commanded line width. You calculate the new flow rate as: (expected wall thickness ÷ measured average thickness) × current flow rate. Most printers work well in the 90–105% range, so if your result falls outside that range, the problem probably lies somewhere other than flow itself.
Extruder e-steps calibration
E-steps is the number of stepper motor steps needed to feed 1 mm of filament — if this value is miscalibrated, no amount of flow rate tweaking will give a stable result, because the error grows proportionally with print length. The test involves marking 100 mm of filament above the extruder, sending a command to feed exactly 100 mm, then measuring how much was actually fed. We cover the full step-by-step procedure, along with the formula for the new e-steps value, in our guide to extruder e-steps calibration.
If under-extrusion still shows up only at higher speeds after calibrating flow and e-steps, go back to temperature — too-low nozzle temperature at high volumetric flow is one of the most commonly overlooked causes, because it only shows up above a certain print speed and looks fine on slow test prints.
How to diagnose under-extrusion step by step
- Visually check the filament segment coming out of the extruder — flattening and gouging point to an idler problem.
- Run the e-steps test (100 mm command) — if the result is off by more than 2–3%, fix the calibration before moving further.
- Print a test wall at 100% flow and measure its thickness with calipers.
- If the wall matches the expected thickness but your actual prints still show under-extrusion, suspect a partial nozzle clog that only shows up at higher speed.
- Check whether the filament was stored in a dry box — if not, dry it before further testing.
- Raise nozzle temperature by 5 °C and repeat the test print — if that helps, temperature was the main limiting factor.
- Only at the end, adjust flow rate in the slicer — it's the last step of diagnosis, not the first.
Settings for popular configurations
How a printer feeds filament also depends on its design. Below are typical starting points for the most popular setups.
| Configuration | Typical setting | Note |
|---|---|---|
| Bowden | retraction 4–6 mm, moderate speed | the longer filament path increases under-extrusion risk at high speed |
| Direct drive | retraction 0.5–2 mm | shorter filament path, less prone to retraction-induced under-extrusion |
| CoreXY with enclosure | check the internal enclosure temperature | a hot enclosure can weaken hotend cooling and encourage heat creep |
| Multi-material system (AMS/CFS) | flow rate set per slot | different filaments in the system have different densities and may need separate calibration |
The most common mistakes
Changing flow rate as the first step. Raising flow without checking mechanics masks the symptom instead of removing the cause — the problem comes back with a different filament or speed.
Ignoring the extruder clicking sound. It's the earliest sign of the drive gear slipping — the longer it's ignored, the more the filament gets chewed and the worse the situation gets.
Testing on the target model instead of a simple wall. A complex model introduces too many variables at once — retractions, speed changes, bridges — making it hard to isolate the cause.
Skipping e-steps calibration. Without correct e-steps, any further flow calibration only patches the symptom, not the source of the problem.
Storing filament outside a dry box between print sessions. Even a few days in open, humid air can noticeably reduce effective flow.
Making too big a temperature jump at once. Raising temperature by 15–20 °C at once makes it hard to tell which parameter actually solved the problem — it's better to change it 5 °C at a time and test.
Measuring wall thickness at the corners. Corners are thicker than straight wall sections because of overlapping head movements, so measuring there inflates your flow calibration result.
Frequently asked questions
Does under-extrusion always mean an extruder problem?
No. It just as often comes from a partially clogged nozzle, wet filament, or too-low print temperature — the extruder itself can be perfectly healthy while the problem sits further along the filament path.
How can I quickly tell if it's under-extrusion or something else?
Thin, see-through walls and holes in the top layer despite correct infill are the telltale signs — if you see both symptoms together, under-extrusion is the most likely cause.
Is raising flow rate above 110% safe?
It usually doesn't solve the problem and leads to over-extrusion elsewhere on the model — if a wall test shows a value above 105–110%, look for a mechanical cause instead of raising flow further.
Does every filament need its own flow calibration?
In practice, different materials — and even different colors from the same brand — can vary slightly in density, so it's worth keeping a calibration result per material rather than applying one value universally.
How long does a full under-extrusion diagnosis take?
The e-steps test and the calibration wall together usually take 30–45 minutes, including print time for the samples — not much compared to the hours of wasted printing on models you'd otherwise have to reprint anyway.
Summary
Under-extrusion can almost always be broken down into one of three groups of causes: mechanical (nozzle, idler tension), material-related (moisture, diameter tolerance), and software-related (flow rate, e-steps, temperature) — and diagnosis only works if you check them in that order instead of starting with slicer settings.
If you suspect recurring problems with a specific filament — a wide diameter spread, inconsistent density between spools — it's worth checking out Porima PLA, where diameter tolerance is ±0.03 mm, or browsing our full PLA range — consistent diameter makes it easier to keep flow rate stable between spools.