Can’t support MOSAIC enough. I’ve owned a composite LSA called a Flight Design for about 5 years, been in arguments with a lot of old timers about the safety aspects but I consider revolutionary and far safer than a certified aircraft than the statistics will tell you.
For example it has a ballistic parachute that will bring the entire aircraft to the ground. Unlike the Lycomings and Continentals the engine wasn’t designed in the 1950s. It’s equipped with real time satellite weather, GPS autopilot, Avionics that would cost you $15-20k to put in a Cessna due to all the red tape.
I will get a lot of heat for this but I think the FAA has killed a lot of people. If pilots had low cost access to things like glass cockpits, satellite weather, inexpensive autopilot, and a healthy ecosystem of cheap, modern aircraft with modern engines (Basic things like fuel injection) a lot of pilots might still be alive right now.
> I will get a lot of heat for this but I think the FAA has killed a lot of people.
On the balance, the FAA has saved more lives than it's cost just because big jetliners hold SO MANY PEOPLE.
But otherwise I fully agree with you.
It's deranged that students these days are taught to manually lean the mixture until the engine sputters (i.e. the engine begins to DIE), then bump it back up.
It's incredibly stupid that an aircraft flying in 2025 has multiple solid state accelerometers and gyroscopes on board as part of people's phones - but the only certified one is a vacuum-powered analog instrument from 1981.
And why the hell do we still fuel GA aircraft with a gasoline that's literally ILLEGAL to use anywhere else?
Don't even get me started on the DPE or medical systems.
These changes cannot come soon enough, because the entire GA world has slipped through the cracks as the FAA has become a disaster.
> And why the hell do we still fuel GA aircraft with a gasoline that's literally ILLEGAL to use anywhere else?
Unleaded avgas is a thing now. But it won't work with many legacy aviation engines. I hope this new rule will finally enable some engine (and thus fuel) innovation.
I think it was only in the last 5? years you can no longer buy a brand new helicopter with manual throttle control right on the collective. Not even RPM control, just a stupid manual linkage going straight to the carb :|
The massive effects of lead poisoning are still seen on a large scale near airports and in areas with lots of avgas powered traffic. That most of it is for essentially recreational purposes makes it even more infuriating.
> It's deranged that students these days are taught to manually lean the mixture until the engine sputters (i.e. the engine begins to DIE), then bump it back up.
Why? That is essentially an engine check. I think you want pilots to understand their machines and to do that you need to operate the engines.
I'm not sure I see the value of doing an engine check while flying the plane, or flying carbureted engines without automatic mixture control in general.
Pilots will understand their engines' quirks better, but those quirks are stupid and obviated by newer technology.
Forget mixture, carb ice keeps killing people in 2025. Even just moving everyone to dumb mechanical fuel injection will be a big step forward at this point…
.......attain strait and level flight, reduce airspeed to 65knots.....pull main controls aft till they hit the stops while applying full rudder......bieng able to recover from the "unusual attitude" quicky and cleanly is the skill needed to procede to all of the technological fiddling and gadgetry.....or is this all just a song and dance working up to just eliminating general aviation and making one of the passengers sit up front and wear a nice hat?
if not, then what on earth are all of these fancy accelerometers for?
For flying in the clouds mostly. These air driven gyroscopic contraptions were fine engineering for their time, but trusting one’s life to one of these when we can have a multiply redundant solid state alternative? No thanks.
Many people have died from believing that their satellite weather was real-time. It’s 3-15 minutes delayed typically.
Go up on a VFR day with widely scattered wet clouds and maybe scattered showers or virga. Fly near some of the wetness and notice they are physically quite displaced from where they are on your weather display. In some cases, your weather display will show weather on the left and clear right, while your windshield will show the weather is actually on your right, having moved during that delay.
Imagine then trying to navigate that in more serious weather and you’ll hopefully get religion that the display is definitely not real-time and must not be relied upon as if it were. It’s a strategic tool, not a tactical one.
Doesn't surprise me. I got on a flight that had to go back to gate due to an "iPad failure" about 6 years ago. The captain was apologising as we disembarked. I asked him what happened. He said software issue across both their flight plan iPads. Hardware was fine. Not sure if it grounded other flights but we were back on a different plane about 2 hours later.
To be clear, we're actually talking about two separate things, they both just use an iPad.
You're talking about what's known as an electronic flight bag, which is using a tablet as a replacement for paper maps and charts. This is legal and has been for some time.
What I and Animats are talking about is running an app on a ipad as a "replacement" for equipment or training that is required for safe, legal, operation. NOT legal.
Maybe because the iPad has a more user-friendly interface than most avionics.
Here's the cockpit of the Gripen, Saab's current fighter aircraft.[1] One big screen across the whole panel. No sign of a classic altimeter or compass. There's a HUD as well, so there's a second device for the basic flight instruments.
The worst typical outcome for pilots who get these violations is taking training, getting signed off to that effect, and a “709 ride” to get their certificate back. They might also face a short suspension.
There's actually a self-report line (run by NASA, no less) where pilots can report stuff like this.
It's really quite brilliant.
The way that you get pilots to do it is that if you drop the dime on yourself it actually gives you virtual immunity from consequences, as long as no injuries or serious damage actually occurred. The idea is to actually find systemic process/procedure issues.
I ride with my dad in a Cessna a lot and he emphasizes you should always be prepared for GPS and radio failure, and you must be able to safely get on the ground with nothing but a map, compass, and your eyeballs.
Don't get me started. It's virtually impossible to buy a car without fuel injection nowadays, but most GA airplanes still use carburetors. And these are vehicles that are constantly changing altitude, so carburetors are even more unsuited for airplanes than for cars.
Okay I lied. You can buy fuel injected engines for airplanes. They are readily available. They merely cost 2x or 3x the price of carbureted engines.
I hear you in general, but GA engines need to work more reliably and go through way more intense operational challenges than automotive engines. Picture this: you take off at 100°F on the ground at full throttle with only air cooling to save weight and reduce potential failure points, then climb to 10,000 feet where the outside air temperature drops to around 65°F (or even freezing conditions). The baffling is a bit worn, and the pilot maintains 70 knots in the climb, pushing cylinder head temperatures (CHT) to 420°F or higher. Then the pilot gets chatty and pulls the engine back to near-idle while CHTs are still at 350°F, before pushing the nose down to kill altitude—causing CHTs to plummet to 250°F in minutes. Through all these extreme thermal cycles and temperature swings, the engine simply cannot quit on its pilot.
GA engines may look antiquated—with their carburetors, magnetos, and mechanical fuel pumps—but this apparent simplicity is entirely by design. These “outdated” systems are actually time-tested solutions engineered for ultimate reliability when failure means catastrophe. While car oils use metallic detergent additives, aviation oils must use ashless dispersants to prevent spark plug fouling that could cause engine failure. The oils must handle sustained high RPM operation and brutal temperature cycling while meeting strict FAA specifications that prioritize proven reliability over cutting-edge performance.
Every component, from the dual magneto ignition (no electrical system dependency) to the mechanical engine-driven fuel pump, represents decades of refinement focused on one critical goal: the engine will not quit when you need it most. It’s not that these engines are behind the times—they’re precisely engineered for their mission-critical role where proven, simple systems trump technological sophistication.
Most of what you describe can be accomplished with diesel Jet-A engines. These used to be unthinkable for GA because they were too heavy, but clean-sheet designs are making it possible.
Much easier to fuel, no electrical system dependencies, no spark plugs to foul, liquid cooling to keep the temps more constant, and dual redundant FADECs. Plus much better range.
It is not true that reliability requires old-style engine design, it's more a question of cost. Modern jet airliners (their engines but also really everything about them) have a ton of complexity, including a myriad of electrical control systems, yet they are no less reliable.
It's just that this is not a fair comparison because manufacturers of said airliners have more resources for R&D.
Except now we’re back to one of the main points of the article - modern airliners cost billions of dollars to develop and certify, and GA aircraft will never get that level of investment.
Professional here also introducing an element that's unexpected. We expect that they'll have more training, they've often done simulator training which is more realistic, they have a lot more hours and so on.
But because it's a job they have much less Plan Continuation Bias aka "Get-there-itis". Flying New York to Dallas? I did that yesterday, and the day before, and the day before that. So if the weather looks bad and maybe we shouldn't, well then I guess we just don't go, I'll go tomorrow, or maybe somebody else will, it's just a job.
GA pilots are notorious for this problem, and it puts them in vulnerable situations where they're one problem away from disaster, as weather is worse than they hoped, things don't happen the way they expected, and gradually they go from "It'll probably be fine" to "I hope I live to learn from this experience".
GA engines use low compression ratios and high displacement to generate power at lower RPMs, reducing mechanical stress and heat buildup that would be catastrophic during sustained high-power operations like climb and cruise. The need for low propeller RPM means designs either go for a gearbox-driven high compression, low displacement approach like Rotax, or the low RPM, low compression, high displacement route like Lycoming—and given these constraints, the good old Lycoming design isn’t all that bad.
The comment was not AI generated. I had grammar spelling and language fixed because english is not my first language and culturally the way I express myself does come across as harsh to sensitive American audiences.
Do you believe anything I say is inaccurate or do I need to accommodate another “please don’t do X because otherwise I feel offended/cannot trust a random stranger/…” stance?
Sorry, it is very obviously AI generated. And yes, there are several inaccuracies or misleading statements in those three short lines.
>The oils must handle sustained high RPM operation
Flat out wrong, most GA piston engines are quite low RPM and even the "higher RPM" engines are rated lower than an equivalent car engine. Redlines are lower than car engines too.
>aviation oils must use ashless dispersants to prevent spark plug fouling
Also flat out wrong, lol. FAA allows use of straight mineral oil and although most people break-in with mineral oil and switch to oil with ashless dispersants, the use of straight mineral oil for an engine's entire life is perfectly legal.
>(the oils must meet) strict FAA specifications that prioritize proven reliability over cutting-edge performance
Another lovely LLM hallucination. I would love to see any sort of FAA "specification" on engine oil that causes a serious performance compromise.
The main thrust of your comment is - and I quote - that the use of carburetors instead of fuel injection is "entirely by design." That is entirely bullshit. Fuel injection was not a mature technology until the 80s and didn't even become the default in new passenger cars until the 90s. If you are designing, let's say, the Lycoming O-320 - one of the most popular GA engines today - in the early 1950s, you used a carburetor because it was the only real option.
I say this all as a supporter of old, simple systems, and as a man who has trusted his life to old, reliable, simple engines. I would love a debate about the actual reliability and factors of reliability of GA engines. But I would have that debate with a human. Because, for all their merits and uses, LLMs currently struggle to produce real insight.
> It's virtually impossible to buy a car without fuel injection nowadays, but most GA airplanes still use carburetors.
That's because GA airplanes in general are significantly older than most cars on the road. If you're buying a brand new car and a brand new GA airplane you're equally unlikely to encounter one without fuel injection.
Now the reason this is the case is because instead of buying brand new GA airplanes, most people are just trading around the old ones because new GA airplanes are ridiculously expensive. This is also a regulation problem, among other things, but the fuel injection thing isn't because modern GA is behind the curve. The issue is that the vast majority of GA pilots aren't flying modern GA.
Go to Oshkosh some year for EAA. You'll see that the vast majority of brand new "affordable" GA airplanes have carburetors. Affordable GA airplanes do exist, for some definition of affordable.
We're probably agreeing with each other but using different words.
>far safer than a certified aircraft than the statistics will tell you
I share your frustration with the technological stagnation of general aviation, but this is completely damning. Cirrus added all of the features you mention, at great expense and in a fully certified aircraft, and took decades to show any kind of clear safety advantage over clapped-out Cessnas (as I understand it, the vast majority of improvement came from intensive training in when to deploy the parachute, which was wildly less intuitive than anyone originally realized and likely remains so for pilots without specialized training). Digital instruments, weather displays, and automation have significant benefits for many use cases, but it's unclear that they're inherently safer than legacy systems for amateur aviators.
Not only it took a focused training campaign to get people to use the chute, all the increased training did was take the plane from having some of the worst safety statistics in the first decade to somewhere around average to slightly better than industry average.
A confounding factor here is that when a shiny new "safer airplane" is on the market you know who it attracts? The least safe pilots. All the doctors and dentists bought Cirruses.
Risk compensation is real... they put themselves into marginal situations because they're telling themselves they can always just pop the chute.
> wildly less intuitive than anyone originally realized and likely remains so for pilots without specialized training
AIUI the specific problem was that humans are bad at "calling it" and the parachute isn't magic. If you used the chute on time you're saved, if you spend that time working through all other options which don't save you, then deployed the chute with no time, you're still dead. So the training was to teach people to call it - yes maybe I could restart the engine (but if not I die), maybe I could keep looking and see that state road (but if not I die), however I could pull the chute right now and almost certainly live so I need to make sure I do that before it's too late.
Suppose in a board game you have three choices. One is worthless, we'll lose, one is 80% chance to draw but otherwise lose, one is a utter gamble maybe 5% chance to outright win otherwise lose. Many players will take the 5% chance. In fact in professional sports not taking that risk often annoys fans - they're here for the thrill. But flying an aeroplane isn't a game, the "outright lose" case is you die and if you have passengers they die too. You should take the draw when it's offered, and if we have to train people to do that then I guess that's what it takes.
It's true! I have only ever flown my Cirrus, in which I did ALL my training. I have no idea how to do an engine out landing. They don't teach it. The solution to nearly every major problem is pull the chute.
You can get an instructor to teach you almost anything sensible to be taught in aviation. If the instruction you’ve had up until now didn’t cover it, call up a flight school and ask to be instructed on simulated engine outs.
I can understand (and even agree with) why Cirrus teaches to “pull early and pull often”. It’s not a terrible policy, Cirrus doesn’t exactly suffer financially from a chute pull, and some Cirrus occupants died who could have lived if the chute were pulled earlier or at all.
You might be in for a bad awakening when comparing the reliability and safety statistics of Lycomings to the Rotax engines in Flight Design planes. Even though I entirely share your enthusiasm in general - these “old technologically outdated Lycosaurus engines” are really reliable in comparison…
Rotax engines have been extremely popular in Europe for LSA equivalents - but boy do I recall countless stories of engine failures. The most crazy one was of a flight instructor that had a total of 12 (!?) before he quit flying. A lot has to do I believe with the “creative ways the engine and its components are stuffed into different airframes”.
"Over the 6-year study period between 2009 and 2014, 322 engine failures or malfunctions involving light aircraft were reported to the Australian Transport Safety Bureau (ATSB) and/or Recreational Aviation Australia (RA-Aus). These reports involved single-engine piston aeroplanes up to 800 kg maximum take-off weight.
Aircraft powered by Jabiru engines were involved in the most engine failures or malfunctions with 130 reported over the 6 years. This represents about one in ten aircraft powered by Jabiru engines in the study set having reported an engine failure or malfunction.
Reports from Rotax powered aircraft were the next most common with 87 (one in 36), followed by aircraft with Lycoming (58 – one in 35) and Continental (28 – one in 35) engines.
When factoring in the hours flown for each of these engine manufacturers, aircraft with Jabiru engines had more than double the rate of engine failure or malfunction than any other of the manufacturers in the study set with 3.21 failures per 10,000 hours flown."
(When you read on, it appears the Jabiru engine was improved and now has less failures)
I do not know how widespread Rotax engines are in Australia and how large the GA is there. Also, I do not track standardized failure rates of engine models around the world - but shared anecdotal evidence. Except two instances, ALL reports or stories from friends and acquaintances around engine failures involved Rotax engines (probably 5:1 ratio). Tracking planes with up to 800kg in the study eliminates all Pipers and Cessnas - which I admit I used as a baseline comparison for my statements. I guess the only plane with a Lycoming/Continental engine below 800kg that comes to mind is the Pa18 from the 1940s/1950s.
Now, I definitely do not say that these are bad engines, but there is a lot of chatter in Europe how these engines are plugged into a wide range of airframes and there are more complex system interactions than meets the eye which can cause some problems. Or put differently: C172 and Pa28 are probably among the most common airframes to stuff the Lycomings and Continentals into. I suspect we kind of figured out how to make these work reliably.
Rotax works in MANY many different combinations and many different airframes - so there is that.
> there are more complex system interactions than meets the eye which can cause some problems
I will grant this for sure. Kind of like modern cars though, it's a double-edged sword. On the UAS programme I'm working on it has been absolutely invaluable to be able to just plug into the 912 ECU's CAN bus and gather a ton of engine telemetry (and send it down to the ground for monitoring at the GCS).
Thank you for posting the links and starting the discussion about 912 reliability. I'm going to have to dig into it a little and see if there's any takeaways I need to bring back to my team.
With zero evidence to support this other than my own experience with N=4 of these, I have a suspicion that part of the problem could be that they're not getting sufficient maintenance and inspection because of how simple they are from an O&M perspective and how robust they are in nominal and off-nominal conditions. When I was first working with it and flipping through the operators manual I was kind of shocked to discover that the only real pre-flight actions are: check coolant level, rotate the prop and make sure the oil reservoir burps. There's a startup and warm-up procedure that we follow to the letter but short-term you almost certainly won't notice if you skip it. Before we had our robust telemetry system and checklists in place, we accidentally flew it with only one ECU lane turned on once and didn't notice until we were on the ground. Engine was already off after landing when someone came on the radio and asked "hey guys... in-flight we're supposed to have both lanes A and B on right?" "Yeah..." "The Lane B switch was off when I approached the aircraft...".
To summarize what I'm getting at: this engine, in my experience so far, has a ton of really robust redundancy features and those redundancy features work so well that you may not notice that you've got issues until you've run out of redundancy. I can only think of two situations where we've had issues bad enough that it caused it to "run rough" and trigger a deeper investigation:
- Because our aircraft is unmanned we have electromechanical relays in series with the Lane A/B switches that we can control from the ground both for engine-start safety (the engine can't be started unless both the crew chief and remote pilot have turned on Lanes A and B) and to be able to kill the engine remotely after landing or in an emergency. We had an electrical issue that was causing the relays to chatter, resulting in Lanes A and B getting sporadic power.
- Somehow in one revision of the ever-evolving full-system checklist the "check water separator" item got dropped and no one noticed. It flew probably 10+ flights on that checklist before we had a really rough start, in an environment that was highly conducive to water accumulation in the fuel (large daily OAT and RH swings). We were horrified at how much water came out when I realized that no one had been checking... and yet there had been zero negative effects until there was a big negative effect.
> I will get a lot of heat for this but I think the FAA has killed a lot of people.
It drives my crazy that in 2025 ADSB is still not mandatory for all aircraft. I get there's old timers flying their tail wheels from the 1950s that don't have any electrical components, but this would massively improve GA safety.
Another one is multiplex radio, again, it's 2025, the technology is there. Why are we still seeing so many blocked communications during emergencies in busy airspaces?
I get there's old timers flying their tail wheels from the 1950s that don't have any electrical components
Frustratingly, the FAA hasn't certified a straightforward solution to this problem. Here in the UK, we can use a fully standalone ADS-B transceiver that requires no permanent installation. After a rebate from the CAA, it costs about $500.
I haven't dug too much into the regulations or power requirements around running ADS-B Out purely off a battery but you can get a TSO-certified beacon https://uavionix.com/general-aviation/tailbeaconx/ and pair it with high-capacity LiFePO4 TSO battery https://earthxbatteries.com/product/etx900-tso/ for a not-unreasonable price. I don't know if that would actually be acceptable, but it's a pretty straightforward way to add that functionality to an otherwise steam-gauge-only aircraft.
> Another one is multiplex radio, again, it's 2025, the technology is there. Why are we still seeing so many blocked communications during emergencies in busy airspaces?
The difference between the radio on the WW2 era ex-pilot boat I spent time on recently and the radio in a brand new jet liner is crazy. The Global Maritime Distress Safety System - a global requirement from the 1980s - mandates a digital VHF radio service named DSC - it's not very clever by today's standards and it's hardly the easiest to drive UI - but it's night and day compared to what is provided for aeroplanes. First of all, and most crucial for its core purpose, many crucial elements of a "Mayday" call are automated so that rescuers have the most important information right away even if you're panicked and incoherent, and it won't get "blocked" by low priority callers trying to figure out which gate they're scheduled for, or whether they can get the longer runway for this approach.
It's almost aggressively bad, I guess they couldn't get an OK to use Morse code?
> The only positive method of identifying a VOR is by its Morse Code identification or by the recorded automatic voice identification which is always indicated by use of the word “VOR” following the range's name.
There's such a long way from "Move fast and break things" to "Eh, it was good enough for my grandfather" and it feels as though aeronautics is very close to the latter. This would be OK if the demands on the system were declining or even steady, but they're increasing.
> Unlike the Lycomings and Continentals the engine wasn’t designed in the 1950s.
Rotax 912? :D
I've spent a lot of time getting familiar with that engine over the last two years as part of a large UAS programme. Every time I have to do integration work with it (electrical or CAN) I end up having an even-deeper appreciation of how thoughtful the engineering behind it is.
Edit: also, since people below brought up 100LL... I also deeply appreciate that it runs fine on plain ol' premium mogas. Both because I'd rather not expose my self to tetraethyl lead all the time and because it's really convenient to be able to just load up a few jerry cans on my way out to the field while getting fuel for my truck.
> Avionics that would cost you $15-20k to put in a Cessna due to all the red tape.
A possibly good example of this: I believe that in Switzerland, small planes almost universally carry "FLARM" devices, essentially "TCAS for hobbyists", implemented through unregulated devices from a private company that people just got because they were cheap and useful. Bureaucratic regulation would have most likely killed a project like this.
For example it has a ballistic parachute that will bring the entire aircraft to the ground. Unlike the Lycomings and Continentals the engine wasn’t designed in the 1950s. It’s equipped with real time satellite weather, GPS autopilot, Avionics that would cost you $15-20k to put in a Cessna due to all the red tape.
I will get a lot of heat for this but I think the FAA has killed a lot of people. If pilots had low cost access to things like glass cockpits, satellite weather, inexpensive autopilot, and a healthy ecosystem of cheap, modern aircraft with modern engines (Basic things like fuel injection) a lot of pilots might still be alive right now.