Every FDM print has some texture. The nozzle lays down filament one flat slice at a time, and the edge of each slice is visible up close, that’s how the process works, not a flaw. This guide covers lines that look coarser, rougher, or more uneven than they should. If you’re still deciding what layer height to print at in the first place, for resolution, speed or strength, that trade-off belongs in our guide to layer height in 3D printing instead.
Quick diagnosis
Lines that are visible everywhere but stay even in width and spacing are normal FDM texture. Lines that look ragged or inconsistent from one area to the next point to the causes below. Ripples or bands repeating at the same height all the way up the print are a different, mostly mechanical problem: see our guide to fixing Z-banding and Z-wobble instead.
1. The normal step effect versus a genuine problem
Prusa’s own knowledge base puts it plainly: layer height is the main factor behind both print time and vertical resolution, and taller layers buy a shorter print at the cost of more visible layers. That trade-off is universal to FDM, and a faint ridge pattern you can feel with a fingernail is expected, not a defect.
What separates ordinary layer lines from a genuine problem is consistency. A well-tuned print shows even lines on every wall. A problem print shows lines that vary: rougher on one side than the other, or textured differently than the slicer preview suggested.
2. Layer height and line width working against your nozzle
Lines look coarser than necessary when layer height is too tall for the nozzle in use. Prusa’s documentation caps layer height at under roughly 80 percent of nozzle diameter, so a 0.4 mm nozzle should stay at or below about 0.32 mm, with 0.1 to 0.3 mm covering most everyday prints. Push past that and each layer bulges more to bond with the one below, exaggerating the ridge on the outside wall.
Line width gets ignored more often. OrcaSlicer’s settings documentation recommends starting at 100 percent of nozzle diameter, 105 to 120 percent on outer walls for a cleaner surface, and warns that widths above roughly 150 percent invite blobs and rough surfaces. Swapping nozzles without revisiting line width is a common, easy-to-miss cause.
3. Inconsistent extrusion: temperature, clogs and under-extrusion
If the plastic leaving the nozzle varies even slightly from one layer to the next, the lines will too. Prusa’s knowledge base notes that partial clogs often reveal themselves only mid-print, sometimes after the filament loaded normally, and hotend temperature may need adjusting by 5 to 15 degrees from a generic profile for a specific spool. A hotend running a touch cooler or hotter between layers changes how each line flows, reading as patches that look fatter, thinner, or glossier instead of uniform.
Under-extrusion narrows this further: lines shrink and gaps open between them, so the surface reads as coarser even at a fine layer height. Run a temperature tower and rule out a partial clog first.
4. Mechanical play in the nozzle and gantry
Loose belts and play in the gantry let the nozzle drift slightly off its path, and that drift prints directly into the wall as fuzzy or wavy edges instead of clean ridges. Prusa’s belt tension guidance notes a loose belt commonly shows up as an irregular shape where a circle should be, and recommends pressing each gantry corner against the frame to confirm there’s no play before tightening tensioners evenly.
This differs from Z-banding, where a bent leadscrew or loose coupler produces ripples repeating at one fixed height up the Z axis. Gantry play blurs X and Y motion generally, so it shows on every layer instead.
5. Cooling fan behavior and a first layer that never evened out
The part-cooling fan decides how quickly each new layer solidifies before the next lands on top. A fan that surges, stalls, or runs inconsistently changes that timing layer to layer, and line definition shifts with it: sharp in one section, slumped in another. Dust-clogged blades, a warped duct, or an abrupt fan-speed jump between layers are the usual suspects.
The first layer does something similar from underneath. Prusa’s first-layer calibration process exists because an uneven first layer doesn’t flatten out as the print gets taller. It repeats in every layer above it.
6. Post-processing options once the print is already done
Sometimes the print is finished and reprinting isn’t worth it, so the fix moves to the surface. Sanding works on any filament: start with a coarse grit to knock down ridge tops, then move through progressively finer grits, the same progression Smooth-On recommends after an epoxy coat cures. It’s slow, but it’s the baseline every other method builds on.
A sandable filler primer, sprayed on before a final light sanding pass, fills the valleys between layer lines faster than sanding alone, and works across PLA, PETG, ABS and resin prints.
Acetone vapor smoothing only works on ABS and ASA, since PLA and PETG don’t dissolve in acetone the way ABS does. All3DP’s guide describes suspending the part in a sealed container above high-purity acetone so the vapor swells and re-flows the surface, closing over the layer lines as it re-solidifies. Handle it carefully: acetone’s flash point is around minus 20 degrees Celsius, so it ignites at room temperature, and the Canadian Centre for Occupational Health and Safety lists headaches, dizziness and nausea from inhaling the fumes. Work in a ventilated space, away from any spark, with an organic-vapor respirator and chemical-resistant gloves.
An epoxy coating such as Smooth-On’s XTC-3D works regardless of material: brush it on as a 2:1 mix by volume, let it self-level over about ten minutes of working time, and it cures in roughly four hours to a hard finish you can still sand, prime, or paint, with none of acetone’s fire or inhalation risk.
Layer lines checklist
- Lines are even in width and spacing across every wall.
- Layer height stays under about 80 percent of nozzle diameter.
- Line width is close to 100 to 120 percent of nozzle diameter on outer walls.
- No partial clog, and hotend temperature is confirmed.
- Belts are tensioned evenly and the gantry has no play at the corners.
- The part-cooling fan runs at a steady, unobstructed speed.
- First-layer squish is even across the whole bed.
- Ripples repeating at a fixed height are Z-banding, not this.