3D Print Troubleshooting: Identify the Failure, Find the Cause, Fix the Print

Start from what the failed print looks like, test one cause at a time, and stop changing random slicer settings.

Troubleshoot· · 3 min read

If you change temperature, retraction, speed, and Z together, you will never know what worked. Pick the symptom, run the smallest test that can falsify the top cause, then stop.

Ellis’s troubleshooting index is the community map for named defects (VFAs, thermal drift, first-layer consistency) (Ellis troubleshooting). Teaching Tech covers the slicer-side tests (calibration.html).

How to troubleshoot without random setting changes

  1. Reprint a small known model (cube, first-layer patch), not the 8-hour part.
  2. Write down firmware, nozzle, filament, and which one change you will make.
  3. Prefer a mechanical check (belt, dirt, grind) before a new “magic” temperature.
  4. If the machine cannot repeat a cube, do not tune supports on a miniature.

First-layer failures

Follow the visual table in first printer setup and Ellis’s squish photos (First Layer Squish). Clean the sheet before you assume the probe is dead.

Under-extrusion and inconsistent extrusion

Test If it fails
Hand-extrude / load until a steady bead Path jam, heatbreak gap, cold nozzle
100 mm requested ≈ 100 mm moved (e-steps / rotation_distance) Calibrate extruder (Ellis extruder calibration)
Still thin after e-steps Wet filament, worn nozzle, too-fast volumetric flow

Do not crank flow to 120% to hide a clog.

Stringing and oozing

Stringing on PETG is common even when PLA is clean. Dry the spool first. Only then tune retraction on a stringing tower, after flow and temperature are sane (calibration workflow). There is no universal retraction length — Bowden vs direct, and firmware pressure advance, change the answer.

Warping and corner lift

Usually bed temp + chamber + material, not “more brim forever.”

  • PLA on a clean sheet rarely needs an enclosure.
  • PETG often wants a moderate bed and a clean PEI; too hot can weld to the sheet.
  • ABS/ASA typically need a warm chamber and still warp on a drafty desk.

See PLA vs PETG vs ABS before you buy an enclosure to “fix PLA.”

Layer shifts and mechanical skips

Ellis lists mechanical, electrical, and speed/accel causes (layer shifting). Check pulley set screws and belt tracking before you halve acceleration “because Reddit.”

Ringing/ghosting and vibration artifacts

Ghosting tracks speed and acceleration, loose frame, and input shaping / resonance compensation where the firmware supports it (Klipper INPUT_SHAPER, some vendor “vibration compensation”). Fix loose panels and belts first (Klipper Resonance Compensation).

Z banding / repeating surface artifacts

Repeating every N mm often matches a Z screw or pulley period. Check Z binding, lead-screw debris, and temperature swing. Do not “fix Z banding” with random flow.

Weak layer adhesion and brittle parts

Cold nozzle, wet filament, or cooling too aggressive on a small part. PLA left in a hot car also embrittles after the print. If layers separate only on overhangs, that is cooling/geometry, not “bad PLA.”

Blobs, zits, seams, and surface defects

Seams are a slicer setting (alignment, painting). Blobs after a travel often mean extra restart extra / unretract, or pressure advance unset. Calibrate PA/LA after extrusion multiplier (Ellis pressure advance).

Dimensional accuracy and holes that print undersized

Holes print small because plastic shrinks and because the slicer treats a hole as an inner wall. Use a hole compensation / XY hole setting in the slicer, or design the hole larger. Do not scale the entire part 102% until you have measured a calibration cube.

When to suspect wet filament, hardware, slicer, or model geometry

Clue Suspect
Pops, steam, fuzzy walls on nylon/PETG/spool left open Moisture
Same defect on two different STLs Printer or profile
Defect only on one downloaded model Mesh errors, inverted normals, paper-thin walls
Started after a crash Hardware (maintenance)

Sources