Hands-on testing

How We Test 3D Prints

A setting does not become advice because one print happened to look good. We make a baseline, change one plausible cause, and measure the result. A failed print is still useful: it tells us which explanation did not hold.

Bench equipment and printers

• Prusa MK4Direct-drive, open-frame reference printer
• Creality Ender 3 V2Bowden setup for retraction and extrusion comparisons
• Bambu Lab P1SEnclosed CoreXY printer for chamber-sensitive materials
• Digital calipersFor wall thickness, fit, and dimensional checks
• Contact thermometerFor verifying bed and nozzle-adjacent temperatures
• 0.01 g scale & feeler gaugeFor flow checks and repeatable bed-gap reference

Repeatable test protocol

1. Document the baseline. Record the printer, nozzle, drive system, enclosure state, filament brand and condition, slicer profile, and visible symptom. Photograph the full part and the defect before changing anything.
2. Choose one testable variable. Keep the model, orientation, layer height, nozzle, material, and ambient setup fixed; change only the setting or hardware condition that could reasonably explain the symptom.
3. Print a small proof sample. Use the same purpose-built test geometry or a matched section of the part, then run at least three samples when the result is close or variable.
4. Measure and compare. Check dimensions with digital calipers, mass with the scale when flow is in question, and temperature with the contact thermometer where heat is the suspect. Compare each sample with the baseline, not with memory.
5. Log the outcome. Photograph the result under the same lighting and distance, label the sample with its changed variable, and keep the failed samples beside the successful one. If the change does not reproduce, we report that limit rather than promoting a fix.

What we control, and what we do not claim

Tests are sampled across the direct-drive Prusa MK4, the Bowden Ender 3 V2, and the enclosed Bambu Lab P1S when the mechanism could depend on the machine. We hold the nozzle diameter, filament batch, model, orientation, and slicer version steady for a comparison. One machine can establish a useful observation; it cannot prove a universal setting. The page names the printer and conditions when they matter.

For the standards that govern sources, corrections, and independence, read our Editorial Policy.