A Cozy Mk IV was destroyed in a U.K. crash after a 3D-printed induction elbow failed and caused complete loss of power.
The owner who installed the 3D-printed induction elbow understood from the vendor that it was printed from CF-ABS (carbon fiber—acrylonitrile butadiene styrene) filament material with a glass transition temperature of 105° C (221° F).
The U.K. Air Accidents Investigation Branch measured a glass transition temperature of 52.8° C (127.0° F) and 54.0° C (129.2° F) for two samples of the failed part.
The Air Accidents Investigation Branch report is available here: https://www.gov.uk/aaib-reports/aaib-investigation-to-cozy-mk-iv-g-bylz
The mechanical properties of plastics vary dramatically between materials, and key temperatures are:
- Glass Transition (Tg): Where it starts to soften.
- Heat Deflection: Where the shape will deform under load.
- Melting Point: Where it fully melts.
Unfortunately, the AAIB report does not include chemical testing of the part to determine the type of plastic the failed part was printed from. The measured glass transition temperature is exactly as expected for PLA, and well outside the expected range of ABS, prompting suspicion the part was printed from PLA-CF rather than ABS-CF.
PLA is cheap and useful for prototyping/fitting aircraft parts, but it should never be used in an aircraft interior—and certainly not a critical engine component. I once had a PLA prototype deform due to the sun’s heat on the passenger seat of my car when I made a stop on the way to the hangar.
There is an FAA-approved 3D-printing filament (Ultem), but it currently requires an expensive industrial printer due to the high printing temperatures. Non-critical components can be prototyped on hobbyist 3D printers and test-flown, then replaced with a final Ultem part. For critical components like this air induction duct, the most cost-effective method is prototyping the part with PLA for mechanical fit, then sending the design to a commercial 3D printer to make an Ultem part suitable for flight test.
I would not use ABS for aircraft parts due to toxic smoke when burned, poor UV light resistance, and poor resistance to aromatic hydrocarbons (like found in avgas). My personal choice for non-critical flight test prototypes is PA12-CF, which has excellent chemical and UV light resistance and high temperature resistance. I replace the prototypes with commercially printed Ultem parts when satisfied with the design.
I’m not sure how high temperatures get in the engine compartment of a Cozy, but even 221 F does not seem a high enough failure point for a component being attached directly to the engine.
I was an aerospace advanced materials specialist for most of my career. Many (if not all) materials suppliers were wildly optimistic about the cured physical properties of their products. Always verify!