Diagnosing print failures without guessing

01

Preserve the scene

Before removing the part, photograph the entire build, its orientation, the platform, support, debris, and display or error message. Record the job identifier, model revision, prepared file, machine, material, start time, operator, environment, and any intervention. If safe, note whether the machine is still heating, moving, extruding, exposing, recoating, or reporting a sensor state. Do not reach into active equipment to save evidence.

The location and timing of the first deviation are more informative than the final mess. A detached first layer that later creates a nest of filament differs from a mid-build collision that knocks over a sound base. A consistent horizontal shift suggests a different causal branch from randomly missing extrusion. Prusa’s troubleshooting library distinguishes first-layer issues, under-extrusion, warping, layer shifting, support failure, delamination, and resin-specific failures because each observation narrows the search differently ¹.

• What is visibly different from the intended result?
• Where is the first bad layer, region, or event?
• Did the problem repeat at the same geometry or machine position?
• What changed since the last known-good build?

02

Trace material delivery

Missing or inconsistent material can originate before, inside, or after the delivery system. For filament extrusion, inspect spool drag, tangles, moisture symptoms, drive engagement, heat-zone condition, nozzle restriction, temperature sensing, volumetric demand, and whether the commanded path actually includes the region. A partial obstruction may produce intermittent thin lines; a too-high flow request can resemble a clogged nozzle; and a geometry error can look like under-extrusion when no path was generated.

For resin, powder, paste, wire, or binder, map the equivalent chain: feedstock condition, storage, metering, spreading or pumping, energy or cure action, atmosphere, and waste or return path. Do not apply a filament fix to a photopolymer symptom simply because both created a missing feature. Start with the process mechanism.

03

Trace adhesion, heat, and geometry

Adhesion failures can occur at the platform, between layers, within supports, or at a joined interface. Surface contamination, insufficient contact, excessive or insufficient energy, thermal contraction, peel force, weak support geometry, or an unstable part can contribute. A larger brim may mask a dirty or damaged surface; more exposure may enlarge features and make resin brittle; higher temperature may improve bonding while increasing deformation or emissions.

Geometry creates local process challenges. Tall slender parts amplify vibration and leverage. Abrupt cross-section changes alter thermal mass. Long bridges lack support beneath a span. Closed cavities trap powder or resin. Fine isolated features may cool or overheat differently from a solid region. Review the sliced or exposed representation, not only the original model. Prusa recommends previewing generated G-code and checking geometry warnings when investigating a spaghetti-style failure ².

• Platform condition and setup
• Contact area, orientation, and support stability
• Local heat or cure balance
• Contraction, peel, recoating, or nozzle forces
• Toolpath continuity and model integrity

04

Trace motion and machine health

Layer shifts, repeating surface waves, inconsistent height, collisions, or skew can arise from loose pulleys, belt or screw problems, obstructed movement, lost motor synchronization, frame movement, unstable mounting, electrical interruption, overheating drivers, incorrect limits, or a path beyond the envelope. Inspect safely and compare the artifact with machine coordinates. A flaw occurring at the same machine position across different models suggests another path than a flaw tied to the same model feature.

Do not tune around a mechanical defect. Compensation and lower speed may reduce symptoms while a loose component continues to deteriorate. Follow the manufacturer’s maintenance procedure, verify fasteners and motion transmission with power removed where instructed, and re-establish calibration. NIST’s test-artifact method encourages repeatable measurement of geometric accuracy and surface roughness to assess machine performance after maintenance or recalibration ³.

05

Run a controlled experiment

Write a hypothesis in physical terms: “the first layer detached because surface contamination reduced adhesion” is testable; “the printer hates this model” is not. Choose the lowest-risk observation or experiment that separates competing causes. Clean according to the surface maker’s method and repeat a small adhesion artifact; inspect a preview; substitute a known-good material; measure extrusion mass; or print a standard artifact. Change one main factor when possible and keep all others recorded.

Define success before the test. A visually complete part may still be dimensionally wrong or weak. Measure the feature connected to the failure and repeat enough to distinguish a stable improvement from chance. NIST emphasizes that process variability, accuracy, surface quality, material consistency, and qualification require measurements and reference data, not only visual judgment ⁴. Add the result to the job record so the next failure starts from evidence rather than folklore.

• Observation
• Plausible mechanisms
• Safest discriminating test
• Expected result for each mechanism
• Measured outcome
• Decision and next controlled change