Nobody prints a part hoping it will one day meet a naked flame. But phone mounts end up on hot dashboards, brackets end up behind heaters, and combat robots end up on fire in front of a paying audience. So it is worth knowing: when a 3D print does meet fire, what actually happens?
A recent hands-on burn test put small printed cones of PLA, PETG, ABS, ASA, TPU, PEBA and HIPS in front of a torch, and the differences were bigger than most beginners expect. The PLA family lit easily and kept going, which makes sense given how much carbon is packed into the polymer. What was less predictable is that two PLA cones from two different brands behaved almost like two different plastics. One caught with a dirty yellow flame straight away. The other shrugged off a lighter entirely and needed a blowtorch before it would burn, and then burned hot and clean. High-temperature PLA went further still and put itself out once the flame was pulled away.
The styrene plastics were the worst offenders. ABS, ASA and HIPS all caught quickly and produced thick sooty yellow flames, which is the visual signature of an aromatic plastic burning badly. At the other end, black PETG was reluctant to light at all, and standard TPU refused outright, softening and slumping under the torch rather than igniting. Foam TPU broke that pattern and burned the most aggressively of anything tested, which is a good reminder that trapped air changes everything about how a material behaves.
The practical takeaway is a material-selection rule, not a fireproofing rule. If a part lives anywhere near heat, reach for PETG or solid TPU first. Save PLA for desk toys, prototypes and display pieces where the only heat risk is a sunny window. If your design has to sit near an engine, a stove, a soldering iron or a high-power LED, remember that PLA will warp and sag long before it ever burns, so heat deflection is the number you should be checking, not flammability. And if you must run your own comparison, do it outdoors on a non-combustible surface, in small quantities, with a fire extinguisher within reach and never over a carpet or a workbench.
Try it on your printer. Print one small test cone in each filament you own, label the base with the brand and colour, and store them in a jar. When you next design something heat-adjacent, you will have real evidence from your own spools instead of a guess. Browse heat-friendly PETG and TPU options at flarelab.com and match the material to the job before you hit print.
Frequently asked questions
Is PLA flammable?
Yes. Standard PLA lights easily and keeps burning once lit, because it is a carbon-rich plastic with a low softening point. Some branded PLAs behave very differently from others, and high-temperature PLA (HTPLA) often self-extinguishes once the flame is removed.
Which filament is the safest around heat?
In open-flame testing, PETG and standard TPU were the hardest to ignite. PETG tends to char and shrink away rather than catch, and solid TPU often just melts. Neither is fireproof, but both are better bets than PLA, ABS or ASA for anything near a heat source.
Why do ABS and ASA burn so badly?
Both are styrene-based plastics, and styrene burns hot with a heavy sooty yellow flame. HIPS behaves the same way. That soot is a good visual signal that the smoke is unpleasant, so never test these indoors without ventilation.
Does filament colour or brand really change the result?
It can. Two PLA cones from different brands gave noticeably different results in the same test, because pigments, glitter particles and toughening additives all change how a plastic reacts to heat. Always test the exact spool you plan to use rather than trusting the material name alone.
What should I never 3D print in PLA?
Anything that sits near a stove, heater, engine bay, car dashboard, soldering station or high-power LED. PLA starts to soften well below the temperature where it burns, so it will usually deform and fail long before fire is even a concern.
Reported and rewritten by Flarelab. Original reporting: Hackaday, based on burn testing by Maker's Muse.
Originally published at flarelab.com.
