This is the mechanical equivalent of vibe coding. 3D printing itself isn't exactly to blame but the negligence of the company that created and sold this part and omitted it's use from an inspection.
Just because a part has the shape of an engineered part does not make it compatible, strong, safe, and fit for purpose. This part could have likely been fine if it used a different material such as Ultem.
In what way is this like vibe-coding -- or do you just mean both are bad?
According to the report:
> The aircraft owner who installed the modified fuel system stated that the 3D-printed induction elbow was purchased in the USA at an airshow, and he understood from the vendor that it was printed from CF-ABS (carbon fibre – acrylonitrile butadiene styrene) filament material, with a glass transition temperature3 of 105°C.
I think by vibe coding he means taking these things at face value instead of rigorously looking if they are up to the standard. When coding you would rigorously look if the code is good / produces any bugs. With vibe coding, you give a prompt and just accept the output, which might be full of errors and blow up (or melt). The analogy is that, yes you can print airplane parts, but they were sloppy and just accepted them at face value instead of rigorously looking if they are up to the required (bug free) standard, ie they wont melt.
>I think by vibe coding he means taking these things at face value instead of rigorously looking if they are up to the standard.
Yeah, exactly -- which is why it's a stupid phrase for what happened here.
Not every negligence is somehow equatable to an AI pitfall, it's just on parents' mind so it's the only metaphor that gets applied.
A poorly fit hammer in a world of nails.
I say this as an engineer/proprietor with years of additive manufacturing experience, it's insulting. A poorly chosen and wrongly used process conveys nothing about the underlying fundamentals of the process itself -- it conveys everything about the engineer and the business processes that birthed the problem.
Similarly if I came across a poorly vibe-coded project I wouldn't blame Anthropic/oAI directly -- I would blame the programmer who decided to release such garbage made with such powerful tools..
tl;dr : it's not vibe-coding itself that makes vibe-coding a poor fit to rocket science and brain surgery -- it's the braindead engineer that pushes the code to the THERAC-25 without reading.
I think the idea was that 3D printing made doing a thing accessible, previously required solid fundamental knowledge (and very expensive kit). Now you can just take some specs off the internet and press go.
The comparison does not seem as absurd to me as it does to you. vOv
In the report they tested samples of the part and found that they actually had glass transition temperatures of 52.8°C, and 54.0°C... so sounds like the owner fell victim to false advertising.
The report further states that the part included in the original design (part of the kit) was made of a carbon fiber composite where the epoxy had a listed glass transition temperature of 84⁰C. If there is an element to be critical of along these lines it's that the part as originally designed is supposed to include an aluminum tube at one end that may stiffen the part - the report makes no conclusions whether it truly would have, but notes that the actual glass transition temperature was found to be close to that of the epoxy used in the original design (that is, lower than the listed temperature of the CF-ABS filament)
The person who coined the term vibe coding is now doing a soylent-like [1] experiment where he only will read content that has been regurgitated by an AI [2], so yes I think it's a fair characterization of "vibe coding".
I think a more useful definition of vibe coding is "something you can do when it really doesn't matter". Which requires a hell of a lot of judgement to know when it doesn't.
Installing life-critical parts of shoddy engineering into a vital system of your airplane is a good example of when things do matter.
Taken at face value, this is engineering negligence. I've done industrial design with plastics and 3D printed parts. Regardless of the forming techniques, with plastics you still need to consider properties like minimum melting temperatures, tensile stress, and so forth. Then you must test that rigorously. This is all standard procedure. That information is in the data sheet for the material.
I did a quick search and found that many plastics are governed by ISO 11357 test standard [1]. Some of the plastics I have worked with used this standard.
Also, strictly as a combo 3D-printing and engine enthusiast: Never with a GUN to my head would I install 3D printed parts in a CAR engine, let alone in an aircraft engine. This is spectacularly poor judgement on the part of the owner.
The part was claimed to be ABS-CF. UK AAIB tested the part and found it to have a Tg of approximately 53C. The Tg of ABS is far higher, around 100C. I suspect that the part may have been accidentally printed with PLA-CF (which has a Tg of approximately 55C.)
The original part was fiberglass/epoxy with the epoxy having a Tg of 84C.
HDT does, kind of, but that’s already covered by the load being defined for the various conditions. HDT is always defined at a specific load so it also does not change with load (since load is fixed).
Isn't Tg a poorly defined metric? It seems like thermoplastics will lose their strength as temperature goes up and there's no abrupt transition where there's a near step-change in behavior
Glass-liquid transition temperature, which is approximately where plastics and other materials change from hard and relatively brittle into flexible and rubbery.
As the other comments here noted, it doesn’t exactly mean that the material is safe to use for a rigid part below that temperature, and the transition extends over a range in temperatures, but it does give you a rough idea about the behavior of a material at various temperatures.
I showed this to a pilot friend of mine out of curiosity, he noted that this type of aircraft is usually kit built / home built. So the fact a part of it was 3D printed was not a total shock.
I'm massively paranoid that some cable clips I printed that will sit on a circuit board will perish in the heat. Meanwhile, some idiot couldn't care less about thermal stability for flight hardware!
I wonder if just including the aluminum tube that was effectively acting as a heat break would have been enough...
Really it seems like a problem of not understanding the environment, and testing (with margins) your replacement in it... 3D printing seems nearly entirely unrelated apart from enabling people to make parts.
An injection molded part, for a close more traditional analogue, would presumably have failed the same way here.
Also the glass transition temperature reported in the report is suspiciously low for ABS and the only source on the material is the owner saying the person they bought it from said... I wonder if it was just outright made out of the wrong material by accident.
The difference is, injection molds are expensive. And the type of people who can afford them tend to cover their ass better - or do slightly less insanely dumb things.
3D printing (especially using filament) allows idiots to enter entirely new areas of endeavor.
I agree, but I note that the 3d printing people are making progress in making really cheap injection molds. I wouldn't count on the difference in cost remaining prohibitive enough that only reasonably serious people can afford it for much longer.
Edit: And I hope the lesson that the safety critical people take away from this is "actual engineering work is needed for airplane components" and not "3d printed parts are scary" because sooner or later they'll run into the same issue with parts made in other ways
Also it's insane that they used a bolted joint with plastics on a critical place, the plastic will creep under the clamp load and will lose clamp force.
Well, no, it's in the UK. It also has a gross weight of around 2000lbs, so it's probably not subject to any of the relaxed regulations anywhere, although I don't know how the UK homebuilt rules work these days.
"Two samples from the air induction elbow were subjected to testing, using a heat-flux
differential scanning calorimeter, to determine their glass transition temperature. The
measured glass transition temperature for the first sample was 52.8°C, and 54.0°C for the
second sample"
Yeah, they might have used ABS-CF filament, but unless they got it from a good brand that uses good resin and proper printing parameters, the actual Tg will be lower, plus the stress from the vibration/load could have made the part fail if it was not for the heat later in flight.
Polymaker's ABS is dubious too because it is blended with PETG. They are coming out with a Pro version that has a higher Tg and requires way higher chamber temps to print properly.
The aircraft owner who installed the modified fuel system stated that the 3D-printed induction
elbow was purchased in the USA at an airshow, and he understood from the vendor that it
was printed from CF-ABS (carbon fibre – acrylonitrile butadiene styrene) filament material,
with a glass transition temperature3
of 105°C.
An alternative construction method for the air induction elbow, shown in the Cozy Mk IV
plans, is a lamination of four layers of bi-directional glassfibre cloth with epoxy resin. The
epoxy resin specified for the laminate has a glass transition temperature of 84°C, after the
finished part has been post-cured. The aircraft owner stated that as the glass transition
temperature listed for the CF-ABS material was higher than the epoxy resin, he was satisfied
the component was fit for use in this application when it was installed.
A review of the design of the laminated induction elbow in the Cozy Mk IV plans showed
that it featured a section of thin-walled aluminium tube at the inlet end of the elbow, where
the air filter is attached. The aluminium tube provides a degree of temperature-insensitive
structural support for the inlet end of the elbow. The 3D-printed induction elbow on G-BYLZ
did not include a similar section of aluminium tube at the inlet end.
> An alternative construction method for the air induction elbow, shown in the Cozy Mk IV
plans, is a lamination of four layers of bi-directional glassfibre cloth with epoxy resin. The
epoxy resin specified for the laminate has a glass transition temperature of 84°C, after the
finished part has been post-cured. The aircraft owner stated that as the glass transition
temperature listed for the CF-ABS material was higher than the epoxy resin, he was satisfied
the component was fit for use in this application when it was installed
What a misunderstanding -- glass transition temperature means different things for thermoplastics (i.e. anything that comes out of an FDM printer like the CF-ABS in question) and for thermosetting resins like epoxy that actually undergo molecular cross-linking during the curing phase. Thermoplastics will get soft and can deform without limit, while thermosets get rubbery but still more or less hold their formed shape.
I think an extended quote shows that this was a really bad call:
“ The aircraft owner stated that as the glass transition
temperature listed for the CF-ABS material was higher than the epoxy resin, he was satisfied
the component was fit for use in this application when it was installed.
A review of the design of the laminated induction elbow in the Cozy Mk IV plans showed
that it featured a section of thin-walled aluminium tube at the inlet end of the elbow, where
the air filter is attached. The aluminium tube provides a degree of temperature-insensitive
structural support for the inlet end of the elbow. The 3D-printed induction elbow on G-BYLZ
did not include a similar section of aluminium tube at the inlet end.
Tests and research
Two samples from the air induction elbow were subjected to testing, using a heat-flux
differential scanning calorimeter, to determine their glass transition temperature. The
measured glass transition temperature for the first sample was 52.8°C, and 54.0°C for the
second sample.“
> The epoxy resin specified for the laminate has a glass transition temperature of 84°C
This seems very low for the kinds of epoxy I've used. I wonder if the manufacturer specs are highly conservative? Or maybe the material has a shortened lifespan with even moderate temperatures?
I was thinking about the ABS in the article and wondering if I would have made the same mistake. Close to every car manufactured today has plastic intakes, usually bolted right on top of the engine. The incoming air should help keep it cool, especially on aircraft. Maybe it was the radiant heat from a nearby cylinder that melted it?
There are some incredibly low Tg epoxies out there, such as West Systems 105 where "TG onset" is 54°C and the heat deflection temperature is even lower.
It was an intake manifold, so it's continuously under suction. At the temperatures in an engine bay the plastic probably gradually creeped to a point where the restriction increased the suction and suddenly it collapses completely.
I wonder who installed it. Was the pilot home 3D printing mods for their plane? And is that even allowed? Super concerning if there was a company behind the installation.
I'd think any semi competent engineer would know better.
Edit: from the report - "A modification application was made to the LAA in 2019, by
the aircraft owner2
, to replace the engine’s throttle body fuel injector with a mechanical fuel
injection system. This system consisted of a fuel controller, high-pressure engine-driven
fuel pump, electric auxiliary fuel pump, fuel flow transducer and associated fuel hoses,
filters and fittings. Following flight testing, the modified fuel system was approved by the
LAA in 2022. The modified fuel injection system had accumulated 37 hours in service when
the accident occurred."
So the pilot himself and the LAA were incompetent. LAA is an association for amateur pilots though so I'm not sure what level of rigour they "approve" things with.
Nearly anything is allowed for experimental amateur-built aircraft like the one in this incident. Unapproved modifications to certified aircraft are forbidden in most parts of the world.
Hah! I've actually 3D printed a part of an intake before. Just as a prototype, to allow me to get a Keihin carb on a motorcycle that had a CV carb.
Printed it on an SLA machine though! I was concerned enough about chemical attack even then, even though it was a temporary part. Never really thought about doing it in filament.
At a glance, that looks like worse than merely the negligence of using a new technology.
The whole point of 3D printing is that the material is moldable when hot but rigid when it cools. And people really should be aware that engines get hot.
>The whole point of 3D printing is that the material is moldable when hot but rigid when it cools.
Which means what exactly? Aluminum will go soft under high temperatures as well, yet this part would not have failed if it was made out of aluminum.
The failure is not the material, the failure is someone neglecting the operating conditions or material properties when choosing materials.
This exact part could have also been milled out of some plastic and would have failed the same way. The method to produce that part is only relevant in so far it is open to more people.
I hope 3-D printing becomes obsolete when robots can achieve the same efficiency with using the standard construction materials. That should take away all benefits of 3-D printing over regular builds.
3D printing isn’t entirely about automation, it’s also a way to get shapes that are impossible to manufacture traditionally. Modern rocket engines almost all use 3D printing because the shapes are so highly optimized.
> The aircraft owner who installed the modified fuel system stated that the 3D-printed induction elbow was purchased in the USA at an airshow, and he understood from the vendor that it was printed from CF-ABS
Hardware engineering is hard. Especially for any safety critical component.
In this case engineering was done by someone, who either did not understand the material he was working with, or the operating conditions in which that part was deployed.
Obviously no testing or any kind of proper engineering was done to create requirements validate them and verify them.
Being able to design a 3D model and print it does not mean you are done with engineering. It is just one step in a very long chain, which is needed to produce devices which stand up to their use.
The person that installed should have thought more carefully about it. But the person that printed it and sold it should face some legal repercussions. Totally irresponsible what they did.
correction to the title: the plane crashed because the owner is a moron, not because he bought a 3d printed part but because he failed to ensure his provider is trustworthy and instead used a fly-by-night nobody to fit a machine that can kill him at any moment.
And this is why (at least for the US) aviation parts have such an onerous paperwork overhead, why a seemingly cheap part like a $.50 bolt balloons to much greater. Granted this aircraft was a UK-equivalent to "experimental" in the US, where you can pretty much do anything to it, I'm of the opinion that doesn't excuse maintenance and adding fly-by-night parts that borders on negligence. Stick to a minimum standard, if not just out of shame of something that could happen.
so, if you were thinking "who would use a 3D-printed part", remember that it may otherwise also have been made with some liquid material, but using a mold, and perhaps two parts using a mold that are joined with re-heating etc. - and now it no longer sounds so outlandish.
This is the mechanical equivalent of vibe coding. 3D printing itself isn't exactly to blame but the negligence of the company that created and sold this part and omitted it's use from an inspection.
Just because a part has the shape of an engineered part does not make it compatible, strong, safe, and fit for purpose. This part could have likely been fine if it used a different material such as Ultem.
In what way is this like vibe-coding -- or do you just mean both are bad?
According to the report:
> The aircraft owner who installed the modified fuel system stated that the 3D-printed induction elbow was purchased in the USA at an airshow, and he understood from the vendor that it was printed from CF-ABS (carbon fibre – acrylonitrile butadiene styrene) filament material, with a glass transition temperature3 of 105°C.
https://assets.publishing.service.gov.uk/media/69297a4e345e3...
Isn't this simply a part that shouldn't have been allowed to be sold based on it being both faulty and also misleading?
I think by vibe coding he means taking these things at face value instead of rigorously looking if they are up to the standard. When coding you would rigorously look if the code is good / produces any bugs. With vibe coding, you give a prompt and just accept the output, which might be full of errors and blow up (or melt). The analogy is that, yes you can print airplane parts, but they were sloppy and just accepted them at face value instead of rigorously looking if they are up to the required (bug free) standard, ie they wont melt.
This is more like adding a third party dependency to your project without vetting it.
>I think by vibe coding he means taking these things at face value instead of rigorously looking if they are up to the standard.
Yeah, exactly -- which is why it's a stupid phrase for what happened here.
Not every negligence is somehow equatable to an AI pitfall, it's just on parents' mind so it's the only metaphor that gets applied.
A poorly fit hammer in a world of nails.
I say this as an engineer/proprietor with years of additive manufacturing experience, it's insulting. A poorly chosen and wrongly used process conveys nothing about the underlying fundamentals of the process itself -- it conveys everything about the engineer and the business processes that birthed the problem.
Similarly if I came across a poorly vibe-coded project I wouldn't blame Anthropic/oAI directly -- I would blame the programmer who decided to release such garbage made with such powerful tools..
tl;dr : it's not vibe-coding itself that makes vibe-coding a poor fit to rocket science and brain surgery -- it's the braindead engineer that pushes the code to the THERAC-25 without reading.
I think the idea was that 3D printing made doing a thing accessible, previously required solid fundamental knowledge (and very expensive kit). Now you can just take some specs off the internet and press go.
The comparison does not seem as absurd to me as it does to you. vOv
Experimental Aircraft are less licensed than non-experimental, so this is more of a YOLO pilot.
In the report they tested samples of the part and found that they actually had glass transition temperatures of 52.8°C, and 54.0°C... so sounds like the owner fell victim to false advertising.
Well how confident would you be that this part isn't exposed to temperatures above that glass transition temperature? It is installed near the engine.
The report further states that the part included in the original design (part of the kit) was made of a carbon fiber composite where the epoxy had a listed glass transition temperature of 84⁰C. If there is an element to be critical of along these lines it's that the part as originally designed is supposed to include an aluminum tube at one end that may stiffen the part - the report makes no conclusions whether it truly would have, but notes that the actual glass transition temperature was found to be close to that of the epoxy used in the original design (that is, lower than the listed temperature of the CF-ABS filament)
Installed _on_ an engine that operates at 200ºC!
The implication was that the part took shortcuts and made something that only looked good on the surface. But couldn't stand up to deeper scrutiny.
The person who coined the term vibe coding is now doing a soylent-like [1] experiment where he only will read content that has been regurgitated by an AI [2], so yes I think it's a fair characterization of "vibe coding".
1: https://nymag.com/intelligencer/2014/10/soylent-creator-hack...
2: https://x.com/karpathy/status/1990577951671509438
If vibe coding is shipping code that you don't understand and can't ensure it's safety,
And if this part was simply 3d scanned and printed in whatever material seemed strongest,
Then it could be an apt analogy
I think a more useful definition of vibe coding is "something you can do when it really doesn't matter". Which requires a hell of a lot of judgement to know when it doesn't.
Installing life-critical parts of shoddy engineering into a vital system of your airplane is a good example of when things do matter.
> blame ... the negligence of the company that created and sold this part
That should be so obvious that I wonder if it was DIY by the pilot.
> The Cozy Mk IV light aircraft was destroyed after its plastic air induction elbow, bought at an air show in North America, collapsed.
Vibe coding can make code that is suitable for production. 3d printed plastic can not be a substitute for a fiberglass-metal part.
Taken at face value, this is engineering negligence. I've done industrial design with plastics and 3D printed parts. Regardless of the forming techniques, with plastics you still need to consider properties like minimum melting temperatures, tensile stress, and so forth. Then you must test that rigorously. This is all standard procedure. That information is in the data sheet for the material.
I did a quick search and found that many plastics are governed by ISO 11357 test standard [1]. Some of the plastics I have worked with used this standard.
A spec sheet for that material is here [2].
[1]: https://www.iso.org/standard/83904.html
[2]: https://um-support-files.ultimaker.com/materials/1.75mm/tds/...
Also, strictly as a combo 3D-printing and engine enthusiast: Never with a GUN to my head would I install 3D printed parts in a CAR engine, let alone in an aircraft engine. This is spectacularly poor judgement on the part of the owner.
I wouldn’t be that absolute, but not until Boeing and Airbus use them in their aircraft on a regular basis.
The part was claimed to be ABS-CF. UK AAIB tested the part and found it to have a Tg of approximately 53C. The Tg of ABS is far higher, around 100C. I suspect that the part may have been accidentally printed with PLA-CF (which has a Tg of approximately 55C.)
The original part was fiberglass/epoxy with the epoxy having a Tg of 84C.
Plastics under load have a lower Tg.
Tg does not change with load.
HDT does, kind of, but that’s already covered by the load being defined for the various conditions. HDT is always defined at a specific load so it also does not change with load (since load is fixed).
Isn't Tg a poorly defined metric? It seems like thermoplastics will lose their strength as temperature goes up and there's no abrupt transition where there's a near step-change in behavior
It kind of is, a better metric is HDT (Heat Deflection Temperature), and it is based on curve usually load over temp.
And a datasheet for a (not necessarily the same) CF-ABS filament claims a HDT at 1.82 MPa of 93C: https://um-support-files.ultimaker.com/materials/1.75mm/tds/...
Something funny is going on with this material given the report is saying they measured a glass transition temperature of ~50C.
Tg changes? Or do you mean they deflect sooner under more load?
What's Tg?
Glass-liquid transition temperature, which is approximately where plastics and other materials change from hard and relatively brittle into flexible and rubbery.
As the other comments here noted, it doesn’t exactly mean that the material is safe to use for a rigid part below that temperature, and the transition extends over a range in temperatures, but it does give you a rough idea about the behavior of a material at various temperatures.
Glass transition temperature.
https://en.wikipedia.org/wiki/Glass_transition
Glass transition temperature I think
Sounds plausible but I guess it's something that they would've confirmed, had it been true
Or it was ABS-CF but they forgot to dry the filament /s
I showed this to a pilot friend of mine out of curiosity, he noted that this type of aircraft is usually kit built / home built. So the fact a part of it was 3D printed was not a total shock.
Edit:
https://en.wikipedia.org/wiki/Rutan_VariEze
https://en.wikipedia.org/wiki/Burt_Rutan
Actual report: https://aviation-safety.net/wikibase/487013
Material was CF-ABS
I'm massively paranoid that some cable clips I printed that will sit on a circuit board will perish in the heat. Meanwhile, some idiot couldn't care less about thermal stability for flight hardware!
I wonder if just including the aluminum tube that was effectively acting as a heat break would have been enough...
Really it seems like a problem of not understanding the environment, and testing (with margins) your replacement in it... 3D printing seems nearly entirely unrelated apart from enabling people to make parts.
An injection molded part, for a close more traditional analogue, would presumably have failed the same way here.
Also the glass transition temperature reported in the report is suspiciously low for ABS and the only source on the material is the owner saying the person they bought it from said... I wonder if it was just outright made out of the wrong material by accident.
The difference is, injection molds are expensive. And the type of people who can afford them tend to cover their ass better - or do slightly less insanely dumb things.
3D printing (especially using filament) allows idiots to enter entirely new areas of endeavor.
I agree, but I note that the 3d printing people are making progress in making really cheap injection molds. I wouldn't count on the difference in cost remaining prohibitive enough that only reasonably serious people can afford it for much longer.
Edit: And I hope the lesson that the safety critical people take away from this is "actual engineering work is needed for airplane components" and not "3d printed parts are scary" because sooner or later they'll run into the same issue with parts made in other ways
Is this a Part 103 Ultralight?
Also it's insane that they used a bolted joint with plastics on a critical place, the plastic will creep under the clamp load and will lose clamp force.
> Is this a Part 103 Ultralight?
Well, no, it's in the UK. It also has a gross weight of around 2000lbs, so it's probably not subject to any of the relaxed regulations anywhere, although I don't know how the UK homebuilt rules work these days.
The actual report[1] holds the answer to the question you’re asking.
CF-ABS (or so claimed)
[1] https://www.gov.uk/aaib-reports/aaib-investigation-to-cozy-m...
I wonder what material it was printed with.
edit: It was ABS-CF, which shouldn't be used under stress long-term in higher temperatures than maybe 65-70°C, or lower depending on the blend.
Lower, according to the report
"Two samples from the air induction elbow were subjected to testing, using a heat-flux differential scanning calorimeter, to determine their glass transition temperature. The measured glass transition temperature for the first sample was 52.8°C, and 54.0°C for the second sample"
Yeah, they might have used ABS-CF filament, but unless they got it from a good brand that uses good resin and proper printing parameters, the actual Tg will be lower, plus the stress from the vibration/load could have made the part fail if it was not for the heat later in flight.
Some manufacturers fudge the Tg.
Polymaker Polylite ABS has a claimed Tg of 101°C but the HDT curve clearly shows it starting to lose strength at 50°C, for example.
Polymaker's ABS is dubious too because it is blended with PETG. They are coming out with a Pro version that has a higher Tg and requires way higher chamber temps to print properly.
by the glass transition temperature, i'm willing to bet it was printed with pla (probably pla-cf).
Full report here: https://assets.publishing.service.gov.uk/media/69297a4e345e3...
The aircraft owner who installed the modified fuel system stated that the 3D-printed induction elbow was purchased in the USA at an airshow, and he understood from the vendor that it was printed from CF-ABS (carbon fibre – acrylonitrile butadiene styrene) filament material, with a glass transition temperature3 of 105°C.
An alternative construction method for the air induction elbow, shown in the Cozy Mk IV plans, is a lamination of four layers of bi-directional glassfibre cloth with epoxy resin. The epoxy resin specified for the laminate has a glass transition temperature of 84°C, after the finished part has been post-cured. The aircraft owner stated that as the glass transition temperature listed for the CF-ABS material was higher than the epoxy resin, he was satisfied the component was fit for use in this application when it was installed.
A review of the design of the laminated induction elbow in the Cozy Mk IV plans showed that it featured a section of thin-walled aluminium tube at the inlet end of the elbow, where the air filter is attached. The aluminium tube provides a degree of temperature-insensitive structural support for the inlet end of the elbow. The 3D-printed induction elbow on G-BYLZ did not include a similar section of aluminium tube at the inlet end.
CF-ABS
> An alternative construction method for the air induction elbow, shown in the Cozy Mk IV plans, is a lamination of four layers of bi-directional glassfibre cloth with epoxy resin. The epoxy resin specified for the laminate has a glass transition temperature of 84°C, after the finished part has been post-cured. The aircraft owner stated that as the glass transition temperature listed for the CF-ABS material was higher than the epoxy resin, he was satisfied the component was fit for use in this application when it was installed
https://assets.publishing.service.gov.uk/media/69297a4e345e3...
What a misunderstanding -- glass transition temperature means different things for thermoplastics (i.e. anything that comes out of an FDM printer like the CF-ABS in question) and for thermosetting resins like epoxy that actually undergo molecular cross-linking during the curing phase. Thermoplastics will get soft and can deform without limit, while thermosets get rubbery but still more or less hold their formed shape.
I think an extended quote shows that this was a really bad call:
“ The aircraft owner stated that as the glass transition temperature listed for the CF-ABS material was higher than the epoxy resin, he was satisfied the component was fit for use in this application when it was installed. A review of the design of the laminated induction elbow in the Cozy Mk IV plans showed that it featured a section of thin-walled aluminium tube at the inlet end of the elbow, where the air filter is attached. The aluminium tube provides a degree of temperature-insensitive structural support for the inlet end of the elbow. The 3D-printed induction elbow on G-BYLZ did not include a similar section of aluminium tube at the inlet end. Tests and research Two samples from the air induction elbow were subjected to testing, using a heat-flux differential scanning calorimeter, to determine their glass transition temperature. The measured glass transition temperature for the first sample was 52.8°C, and 54.0°C for the second sample.“
> The epoxy resin specified for the laminate has a glass transition temperature of 84°C
This seems very low for the kinds of epoxy I've used. I wonder if the manufacturer specs are highly conservative? Or maybe the material has a shortened lifespan with even moderate temperatures?
I was thinking about the ABS in the article and wondering if I would have made the same mistake. Close to every car manufactured today has plastic intakes, usually bolted right on top of the engine. The incoming air should help keep it cool, especially on aircraft. Maybe it was the radiant heat from a nearby cylinder that melted it?
There are some incredibly low Tg epoxies out there, such as West Systems 105 where "TG onset" is 54°C and the heat deflection temperature is even lower.
> The Cozy Mk IV light aircraft was destroyed after its plastic air induction elbow, bought at an air show in North America, collapsed.
Given what it was would it have been under actual stress?
Certainly seems questionable to use any 3-D printed plastic material for exhaust. That’s absolutely going to be too hot.
It was an intake manifold, so it's continuously under suction. At the temperatures in an engine bay the plastic probably gradually creeped to a point where the restriction increased the suction and suddenly it collapses completely.
I wonder who installed it. Was the pilot home 3D printing mods for their plane? And is that even allowed? Super concerning if there was a company behind the installation.
I'd think any semi competent engineer would know better.
Edit: from the report - "A modification application was made to the LAA in 2019, by the aircraft owner2 , to replace the engine’s throttle body fuel injector with a mechanical fuel injection system. This system consisted of a fuel controller, high-pressure engine-driven fuel pump, electric auxiliary fuel pump, fuel flow transducer and associated fuel hoses, filters and fittings. Following flight testing, the modified fuel system was approved by the LAA in 2022. The modified fuel injection system had accumulated 37 hours in service when the accident occurred."
So the pilot himself and the LAA were incompetent. LAA is an association for amateur pilots though so I'm not sure what level of rigour they "approve" things with.
In addition to what other users have mentioned, the airplane changed hands in August 2024, after the modifications were made.
Nearly anything is allowed for experimental amateur-built aircraft like the one in this incident. Unapproved modifications to certified aircraft are forbidden in most parts of the world.
The plane is an experimental class, so I doubt they have to follow a lot of regulations.
Hah! I've actually 3D printed a part of an intake before. Just as a prototype, to allow me to get a Keihin carb on a motorcycle that had a CV carb.
Printed it on an SLA machine though! I was concerned enough about chemical attack even then, even though it was a temporary part. Never really thought about doing it in filament.
At a glance, that looks like worse than merely the negligence of using a new technology.
The whole point of 3D printing is that the material is moldable when hot but rigid when it cools. And people really should be aware that engines get hot.
I think there's some nuance missing here. "Hot" is a scale, not just a true/false check.
Apparently they thought it's ok because the published glass transition temp is higher than the epoxy used for fiberglass construction
>The whole point of 3D printing is that the material is moldable when hot but rigid when it cools.
Which means what exactly? Aluminum will go soft under high temperatures as well, yet this part would not have failed if it was made out of aluminum.
The failure is not the material, the failure is someone neglecting the operating conditions or material properties when choosing materials.
This exact part could have also been milled out of some plastic and would have failed the same way. The method to produce that part is only relevant in so far it is open to more people.
Bought it at a get-together.
Like gunshows, it’s a magnet for bad ideas.
I think the main issue is that many filament manufacturers mislead or outright lie about their filament capabilities.
I hope 3-D printing becomes obsolete when robots can achieve the same efficiency with using the standard construction materials. That should take away all benefits of 3-D printing over regular builds.
3D printing isn’t entirely about automation, it’s also a way to get shapes that are impossible to manufacture traditionally. Modern rocket engines almost all use 3D printing because the shapes are so highly optimized.
It's a light aircraft, the owner probably built it and is allowed to fix it. So it's probably not a company that printed the part.
Per the report
> The aircraft owner who installed the modified fuel system stated that the 3D-printed induction elbow was purchased in the USA at an airshow, and he understood from the vendor that it was printed from CF-ABS
i'm confused: if they were on final and lost power, why not just glide to the runway??
edit: nvm, i found my answer in the actual report.
Hardware engineering is hard. Especially for any safety critical component.
In this case engineering was done by someone, who either did not understand the material he was working with, or the operating conditions in which that part was deployed.
Obviously no testing or any kind of proper engineering was done to create requirements validate them and verify them.
Being able to design a 3D model and print it does not mean you are done with engineering. It is just one step in a very long chain, which is needed to produce devices which stand up to their use.
The person that installed should have thought more carefully about it. But the person that printed it and sold it should face some legal repercussions. Totally irresponsible what they did.
correction to the title: the plane crashed because the owner is a moron, not because he bought a 3d printed part but because he failed to ensure his provider is trustworthy and instead used a fly-by-night nobody to fit a machine that can kill him at any moment.
Absurd what people will do to save a buck.
Wow. It's called "thermoplastic" for a reason.
And this is why (at least for the US) aviation parts have such an onerous paperwork overhead, why a seemingly cheap part like a $.50 bolt balloons to much greater. Granted this aircraft was a UK-equivalent to "experimental" in the US, where you can pretty much do anything to it, I'm of the opinion that doesn't excuse maintenance and adding fly-by-night parts that borders on negligence. Stick to a minimum standard, if not just out of shame of something that could happen.
The part was a "plastic air induction elbow", i.e. this kind of thing:
https://duckduckgo.com/?q=plastic+air+induction+elbow&ia=ima...
so, if you were thinking "who would use a 3D-printed part", remember that it may otherwise also have been made with some liquid material, but using a mold, and perhaps two parts using a mold that are joined with re-heating etc. - and now it no longer sounds so outlandish.
The picture of the collapsed one is at https://ichef.bbci.co.uk/news/1024/cpsprodpb/7fee/live/52acf...
It would be curious to know what parts and connectors it should look like are.
And that texture on the right hand side of the image doesn't exactly look like something in a healthy engine.
3D printing parts is FAA approved?
Lots of F1 parts are 3D printed, as well as many satellite rocket parts still flying. You just need the proper materials.
Yes. I mean not this one, but the FAA has definitely approved 3D printed parts. SpaceX's raptor rocket engines, for instance.
I'm not sure Gloucestershire is under FAA jurisdiction.
Some of the most advanced aircraft engines for commercial airliners contain 3D printed parts: https://www.cfmaeroengines.com/leap
The FAA denying approval to parts based on how it was manufactured and not how it performed under testing would be totally ridiculous.
Crash occurred in UK
You may want to ask about the LAA: Light Aircraft Association.
This might be Darwin Award eligible!
> The sole occupant was taken to hospital with minor injuries.