Hacker News Reader

Comment by looping__lui | Hacker News Reader

I love innovation and stuff. Radical ideas.

But the aviation market is one of the most heavily regulated - bureaucracy beyond comprehension.

Did you ever wonder why Piper or Cessna airplanes look EXACTLY what they did 50 years ago? And why the engines used in these planes (e.g., Lycoming) are referred to as “Lycosaurus”!?

If you go for a sightseeing flight with a local aeroclub - you will find the pilot spending 30min pre-checking the aircraft, checking the weather, reading NOTAMS etc. Not to mention the potentially pretty intensive communication with ATC et al. required to make sure everybody stays safe.

Getting a pilot license is magnitudes more work than getting a drivers license, proficiency has to be continuously demonstrated, maintaining airworthiness of an airplane isn’t exactly cheap either and pretty heavily regulated.

And all that should be “automated”, certified and approved?

Not saying things can’t be automated - but no shit, the spark plugs in a Cessna are like 50$ each for that “aviation certificate”...

Even if there ARE rules and guidelines how to certify autonomous vehicles like that - like how does anybody imagine that a novel aerial vehicle like this actually IS CERTIFIED within a lifetime?

Pilots still walk around a multi-million $ fighter jet or aircraft equipped with the most sophisticated avionics because “a bird nest in the engine intake is hard to detect and difficult to resolve mid flight”.

Investing in one of the most heavily regulated, difficult to scale and extremely expensive to operate industries is brave... Even more so when this industry is low margin and “kept dying every couple of years”...

View on HN

Private pilot here. Yes, aviation innovates slowly, but it's not nearly as bad as you make it sound. Cirrus aircraft have been a game changer, as has the Garmin G1000 avionics. I can fly just about any instrument approach entirely on autopilot. The only thing I have to touch is the throttle and the flaps. I get real-time traffic information via ADS-B, real-time weather via Nexrad. I do my flight planning in five minutes using Foreflight on an iPad. It is all pretty awesomely cool actually.

It's not installed in the plane I fly, but Garmin's auto-landing system was recently certified for emergency use.

https://generalaviationnews.com/2020/05/19/garmin-autoland-c...

Yes, but let us not compare a 50k used Cessna to a 500k Cirrus?

They rely on a Continental that was developed in the 80s (so quite new) - but doN’t you still have to adjust the mixture manually (e.g., manually adjust the air to fuel ratio)?

I mean, the cool stuff you are describing can easily end up at 500k-2MM USD with a fully trained private pilot...

>let us not compare a 50k used Cessna to a 500k Cirrus?

why not? the original post was that innovation can't happen in aviation because of regulation. if we're shifting the argument to innovation doesn't happen because it's expensive, that's a much more solveable problem.

Let me rephrase: The innovation you describe (which is cool and needs certification) bumped the price for a new airplane from about 50k in the 70s (I guess we may be talking 200k in today’s money) to close to 1MM USD.

And that innovation is “just” glass cockpit, and some more advanced avionics. Not changing the principles - you fly a bit faster, a bit safer but still need a license and do all the preparation and checks. But it’s a bit more fancy. From 200k to almost 1 MM.

The Garmin Autoland you described bumps up the sum to 3MM (https://www.piper.com/press-releases/piper-announces-new-m60...).

So factor 10 for something “that was already unreasonably expensive”...

My argument was around “innovation has a horrible ROI” in the aviation industry.

Building a startup in that space has to take costs into consideration... I mean, Musk initially didn’t shoot people into space, there was an inefficient market and the margins were enormous... That’s a different story...

> “innovation has a horrible ROI”

This may be so, but I don't think you've really substantiated it; you're just talking about part of the "I", without any reference to the dollars of "R". Sure, $1m puts a plane out of almost everybody's recreation budget. But for a piece of transportation infrastructure with high utilization, it might not be prohibitive. (I mean, a jumbo jet costs hundreds of millions to build, involves significant R&D budgets on each new model, and is still positive ROI, just.)

You said "you fly a bit faster, a bit safer" in a dismissive way, but how much is this actually worth? I could easily believe that making flights slightly safer is worth an extra $1m of capex; I think your original claim would be much stronger if you provided some more concrete numbers/analysis here.

What kind of analysis are you looking for? How much these features are worth is subjective and the general aviation market is super small and not exactly growing. Don’t get me wrong: everybody I know loves his SR20/22.

My comment was not dismissive of the innovation by Cirrus (which I think is great).

It’s about the costs getting it into the market and having it certified.

Imagine you buy a car for 50k USD. Now, you want an airbag and a better infotainment system. Not a problem. Price is now 250k. Is an airbag and a better Infotainment system worth that premium to you? I can’t answer that for you. But this is what we are dealing with in general aviation.

And it is not 1MM USD in Capex... It is 1 MM USD premium per PLANE.

And again: there is a reason why you have the B747 and 737 for like 40 years... Because exactly of that certification nightmare of new airplanes.

Now talk autonomous electric vehicles with passengers. How should the math ever add up?

> How much these features are worth is subjective

Perhaps we're talking at cross purposes here; when talking about ROI, the numerator and denominator are usually both in dollars.

If I'm pitching Lilium to some VCs, I'd model out the cost of researching, developing, and building the fleet ("I"), and the revenue I hope to generate in a given timeframe ("R"). After adjusting for the risk of failure, if that ROI is higher than what you can get with other investments, the ROI is, in the relative sense "good". If it's worse than putting your dollars into treasury bonds, it's "bad".

Paraphrasing, your arguments earlier in this thread were that investing in aviation is brave, because the ROI on those investments is bad, because it's expensive to certify new inventions. I certainly agree that it's expensive to build a plane, or even add features to a plane, and even more expensive to build and certify a new plane. But that doesn't mean it's bad ROI. The return generated by building expensive things can be astronomical; see military and aerospace contractors taking in multi-billion dollar contracts with very healthy profits, for example. There should be clear numbers behind this; how many aviation startups succeeded in the last N years? How many folded? That would be the objective analysis I'm looking for when talking about ROI on investments in aviation companies.

Maybe that's not the point you were intending to make; perhaps you're just talking about the ROI for an individual buying a plane, rather than a business operating a fleet (like Lilium in the OP)? But if that's what you mean by ROI, then I don't think that is particularly relevant to investors; they are interested in ROI in the sense I outlined above.

If you are in the aviation industry and building planes - you compete with other means of transportation; e.g., cars, trains, Ubers etc. It is a competitive market with relatively clear revenue expectations per person-miles (e.g., private helicopters potentially being an upper boundary). So, the R side of the equation is pretty known... Helicopter taxi services didn’t quite take off and I guess it is because they were kind a above the per person-mile price people were willing to pay...

You will also be in one of the most heavily regulated industries and push for a product that doesn’t even yet know its regulations...

You can’t compare it to the military because - well, they don’t have to certify the same way civil aviation has to. They can essentially say “it’s good enough”. And they aren’t bearing the “R” side risk of the investment for being expose a free market. Military contractors aren’t “paid by the price of a bomb dropped” in a competitive market.

The R is limited and known. We know helicopter taxi services don’t work. The I is a complete unknown and potentially magnitudes higher than expected. Why would I invest in this market that has historically had a bad track record for innovation?

E.g., you want an airbag in your car and better infotainment? Price for the car goes up from 50k to 250k. In the aviation industry, a Tesla would probably be closer to 500k-750k given all the innovation it has (and would need aviation certification) compared to a 50k “regular car”.

If you think this would not “make you think twice” - ok.

One thing to keep in mind is economies of scale:

Cirrus: "new aircraft deliveries for the piston SR Series reached 380 in 2018"

By way of comparison, Ferrari delivered 9,200 cars in the same year.

Part of the problem with aviation has been that aircraft are boutique items and priced accordingly.

I'd imagine even a Toyota Camry would cost $150k if it was a one-off design.

You only pay the R&D costs once. All you're saying is that you have to put a lot into R&D, but that's all on the I side of ROI. It was the same with SpaceX, the I part was huge. The question is whether you can make the R part work. And if you have scale, then the R part moves differently.

It's the same with the first car you build. The first Tesla Roadster cost tens of millions to make. The second Tesla Roadster cost half that.

Well, the R side pops up when they optimistically tell you that your 20 min commute is going to cost 70 USD...

SpaceX is exactly the WORST example in this context: they just went ahead without all the bureaucratic overhead you find at NASA; they weren’t exactly transporting people at the beginning but cargo... And they had a pretty clear market with fat margins; those margins were fat due to the (unnecessary) overhead other agencies created.

You can’t do that in this case.

Put a bit differently: you can probably guess the magnitude of costs for a car to develop. In this case, I would be more than surprised if the get within a 10x magnitude in the end for their cost estimate and EVER get it certified. Again: there are currently NO guidelines how to certify that...

An example from Airbus and the A400 flagship:

https://en.m.wikipedia.org/wiki/Europrop_TP400

“Several technical problems delayed the engine's certification test program and pushed the entire A400M aircraft program into further scheduling adjustments. The engine delays were primarily due to problems with completing the full authority digital engine control (FADEC) software to the satisfaction of the civil authorities. More specifically, Europrop determined in mid-2008 that the engine worked correctly, but the FADEC software still did not meet EASA requirements.[27] Since the A400M was intended for humanitarian missions, the aircraft also needed to have a civil certification. Europrop did not realize that this meant the FADEC also had to show traceability and accessibility, so EASA denied civil certification of the software. Because of this problem, the first A400M test aircraft, which was flight-ready by September 2008, was not permitted to fly. Europrop had to triple the size of its workforce to fix the issue,[28] resulting in a FADEC system consisting of over 275,000 lines of code, which was four times more complex than the FADEC software for the largest civil jet engine.[29] Other problems included numerous plane subsystems providing insufficient logging to the main aircraft computer.[30]”

That's a $250 flight which is mostly labour turned into a $70 flight without the labour. That's a marginal cost thing. It goes in a different part of the book.

I'm not saying this is doable but those numbers show nothing.

Pilots essentially work for free... It has become “almost industry standards” that young pilots PAY to fly. If I threw a wild guess at 30k USD salaries for these kind of “pilot jobs” with 1000h of flying per year, that puts labor costs to 30-50 USD/h. Maintenance and gas will by far outweigh that.

Helicopter pilots in the Bay Area get paid $30k/year? Okay, I don't know that to not be the case but it sounds hard to believe.

OP might be talking about europe:

https://www.dw.com/en/being-a-pilot-isnt-what-it-used-to-be/...

https://dutchpilotgirl.com/pilot-salary/

No, I wasn’t talking about Europe...

I don’t want to “shame” anybody in particular, but look up your local charter airlines and what they offer...

This can of course vary if there are rather complex requirements (e.g., a certain type rating, experience level etc.) - but being a pilot isn’t cool financially anymore. A small fraction ends up with “the large airlines” - but if you don’t, you are not in for the big money...

I think we have different mindsets here: !every! company in aviation, the automotive or transportation industry is worried about the “I” for the R&D.

If there is a factor 10 whatever uncertainty in this equation - you are essentially just gambling like an Options-Day-Trader on Reddit. I understand that investors see “the hockeystick” unlimited revenue upside and justify everything around that.

But I personally would not want to so business in the most over-regulated, slow-growing and low-margin industry with a “factor 10 investment uncertainty”; that uncertainty btw renders your investment null, void and lost if your plane doesn’t get certified. And you don’t even have certification guidelines.

Are you in the aviation industry? I am not. So know that what I'm saying is based on low-grade near-zero prior knowledge. Would love to know the certainty on your knowledge.

I think what matters more is the line of my argumentation. If I was a long-standing aviation industry veteran I would probably not be the source of great innovation and thus be blind for radical disruptions. Automotive and space disruptors had little to no prior knowledge but asked the right questions and had the right arguments... My line or argumentation does not require much aviation expertise: a heavily regulated market with many uncertainties that put the chance of your investment succeeding at a low percentage, require ALL the investment be made before you can test it on the market is a big red flag; even more when the industry is known for very high costs in the innovation space and you are “disrupting” it; add to that that the return on investments have historically been horrible -e.g., google Cessna, Piper, Beechcraft and a dozen other big players that went bust or were very close to it...

Your arguments can be independently evaluated. Your claim to facts are questionable if they aren't from knowledge. I'll take it that you aren't in the industry and have priors similar to mine on your knowledge.

Thanks for the discussion.

I have been flying planes for 20 years. I own a plane. I worked for the largest commercial airline maintenance company in my country.

> And that innovation is “just” glass cockpit, and some more advanced avionics. Not changing the principles - you fly a bit faster, a bit safer but still need a license and do all the preparation and checks. But it’s a bit more fancy. From 200k to almost 1 MM.

Nah. Avionics, etc, and what goes into a plane is cheap compared to the purchase price. Even when you talk about G1000, G3000, etc, they are not the primary cost drivers.

What's killed the price of general aviation aircraft is that very few units are purchased. In turn, prices go up; in turn more old planes are kept flying and fewer units are purchased.

Four seat aircraft without glass cockpits and without notable innovation are a few hundred K.

There's a lot of innovation now happening in the experimental/amateur-built category. Super-capable avionics, GPS everywhere, electronic ignitions and advanced engine intelligence/management. And for reasonable cost (relative to a new Cirrus and even a new Cessna). If you're willing to do the building yourself or trust another builder, you can get a very capable, modern airplane for under $100K that puts the certificated 1970s Cessna to shame in every measurable dimension.

I think what you are stating should be a top comment and shows what the certification and paperwork overhead do in terms of “purchasing costs”. I doubt these planes are less safe (but I admittedly don’t have the numbers ready) - but they are a fraction of “certified planes”

A new Cessna 172 SP with a Garmin G1000 glass cockpit and 180HP engine is 230,000$ that’s a significant upgrade over the 1970’s model for similar costs. Or you can spend 1/2 that on a The Cessna 162 Skycatcher with a 100HP engine and Garmin G300 Avionics.

100+k is expensive, but even a 162 is likely to be flying for 40+ years.

Yes, you do have to manually adjust the mixture. I have to admit, for the newer turbo cirruses its insanely easy to do. just have to pull it until your fuel flow matches a blue indicator on the display.

Private pilot, instrument rated here ;). I fly a G5 SR22 Turbo, which is state of the art for GA piston aviation, or as the rest of the calls it, state of the are for 1982.

No autothrottle, no FADEC, and that plane new costs close to $1 mil. Short of the turbine world, its the best you can get but its still ancient tech. Following a magenta line on autopilot is not a hard problem to solve.

Avionics have certainly changed a lot, but basic aircraft structures and controls have been almost stagnant for decades (and most of the places they aren't static are largely related to electronics--even if you're lucky enough to have a FADEC running your engine, it's still controlling hardware that largely dates to the '50s or so).

That is super cool! I always wonder why we don’t have something like this if we hear a story about a nonresponsive aircraft eventually running out of fuel and flying into a mountain.

Automation cuts both ways. It can kill you when it doesn't work right just as easily as it can save you when it does.

Lots of innovation has happened in aircraft electronics. The powerplant? Not so much. Most light aircraft engine designs are four or more decades old, and the only ones cheap enough for a moderately-priced aircraft have carburetors. I don't think you can even buy a car these days with a carburetor, but with airplanes they're still standard, even though fuel injection makes a lot more sense in an airplane than in a car. You can get airplanes with fuel-injected engines, but a fuel-injected Lycoming IO-390 engine (like you'd find in a Cirrus) will cost about $27,000. Used. Just for the engine. You can get a decent car for $27,000, and it will absolutely have a fuel-injected engine. Likely a turbocharger as well. And the car's engine will produce a lot more horsepower.

Oh and you'll have to fuel the airplane with avgas which costs $5/gallon and contains lead, and you have to wonder WTF??

There's no technical reason aircraft engines cannot be turbocharged, fuel-injected, fueled with regular unleaded gas, and cheap. Aircraft engines should have followed the innovation advancements that have happened in car engines, but they haven't. This is probably because of regulation, monopoly power, liability, and a host of other reasons, but as an engineer I find the situation ridiculous.

Car horsepower and airplane horsepower are very different things.

A car engine has to be able to maintain peak power output for a few seconds while accelerating. maybe a few seconds, unless you're towing a load up a hill or something. On level ground you're using a small but variable amount of your available power.

An airplane engine has to maintain peak power for the entire climb-out, and a substantial fraction of that for cruising for hours at a time.

unless you're towing a load up a hill or something

This reminds me of how a lot of people are surprised when they find out the engine of a typical semi-truck has "only" a few hundred HP but several times the displacement of a passenger car engine rated at almost the same power. The car engine will wear out quickly at continuous full load, whereas the semi engine is designed to run at full load continuously. Ditto for engines driving generators and pumps --- full load is normal load.

US$5/gallon is about the same price as our vanilla automobile juice here in NZ.

We pick up avgas (half the price) from the small airport near us if we’re doing strictly off-road activities in our vehicles (no fuel tax).

Interesting. I had assumed avgas was more expensive than auto gas everywhere but I guess not.

Fuel in NZ is $5 NZD? Has Covid had an impact on that local price or is that standard? It's

Less than $1.50 AUD across the ditch at the moment. And that's a bad day.

Per gallon. $1.30 / L => $5.85 / gallon.

Petrol is/was always more expensive in NZ though, around ~70c/L from memory. On the flip side, registration and insurance is much cheaper.

Ah, thanks! I missed the measurement (per galling to a litre). That makes more sense.

And cheaper than car gas here in Europe (Lithuania).

Reminds me of Germany in (I guess) the 80s, where police was sometimes checking diesel cars on whether owners had used (lower taxed) heating oil for their cars (which is technically tax evasion). It might still be a thing.

In eastern Europe(Baltics) it is still a thing because people just bring over Diesel from Belarus or Russia by cars with big fuel tanks and low fuel consumption, so they simply cross the border and at the first fuel stop brim the tanks and go back to europe, In Belarus it is especially easy as the fuel price is mandated by government and is equal across the whole country. And police sometimes checks the fuel tank for non taxed fuel, it is either Green(Agricultural, illegal for anything else), Normal taxed, and Russian(If you can prove that it was "imported" by you then its ok) non taxed.

They still test it a lot actually in Germany. Especially on trucks. Using heating oil instead of the diesel is a big cost advantage...

There's no technical reason light aircraft engines can't be a heck of a lot better than turbocharged, fuel-injected pistons. They could all be cheap turbojets designed for low altitude flight burning regular diesel at 10gph going 300 Knots, with far higher reliability than any piston. Unfortunately for the reasons you mentioned, and the myth that turbines can't be efficient at low altitudes, nobody is going to develop that engine.

Eclipse tried it and went bankrupt. But it was a nice idea.

https://en.m.wikipedia.org/wiki/Eclipse_400

The problem with the 100LL phase-out is that you have over 70 years of engines and fuel systems designed to work with leaded fuel.

In order to complete the phase-out you either need to: force people to put a new $30K+ engine into a <$50K plane, totally replace the fuel system, and get a supplemental type certificate or: Require that the new fuel is a drop in replacement. One would have to show that the new fuel is 100% compatible with 100LL without modifications to the aircraft.

The FAA has been attempting the second route for some years, but to date has not been terribly successful. The first route has been done successfully in other regions such as the EU, but I would suspect there is far less legacy general aviation aircraft. I personally look forward to the mainstream introduction of diesel engines into the US market that are able to run on Jet-A

You have a sensible third option which is how automotive changed for the most part.

Grandfather in current planes and mandate new models to use new fuel only. Over time, perhaps lot longer than cars(planes do have longer life) you will have changed all of them.

Had they started it in 80's and 90's same time as auto-mobiles, most planes would have changed by now.

Every industry faces the same challenges to transform technology, they are able to move forward without as much difficulty as aviation ?

No they wouldn't.

You're underestimating the age of the fleet.

The average GA aircraft is 50 years old, and that is likely to increase in the near-medium term as production of GA aircraft is practically zero.

Source: https://www.ainonline.com/aviation-news/business-aviation/20....

(Cirrus, mentioned above, produces about 300 aircraft a year - compare to the 100,000+ Ceessnas on GA fields, almost all of which are over 40 years old.

1980 is 40 years ago now . In another 10 years half the fleet would have changed .. still better than we are now .

You could also speed it up by making costlier to run older ones over time . Keeps the industry more active too..

You're assuming production is constant - it isn't. It vastly dropped off in the early 80s due to increasing liability costs, and has never recovered.

Well, one factor would be, that people are happy to fly old planes. If due to the environment regulations, you would have to buy more new planes, demand and consequently production would increase.

That would have to be onerous regulation.

The cost difference between new and used is truly immense.

A new Cessna 172 starts at around $400,000, and that's before you start checking off things on the options list.

Used can be as low as $20,000 for examples for the 1960s. If you thought depreciation on luxury cars was bad...

The GA system is already frankly dieing due to ever increasing costs. Your suggestion would just kill it for good.

I was working on my Private Pilot license in the mid-90's. I quit just around 40 hours with my only remaining required training maneuver being a Night cross country.

I quit for pretty much the reasons you give: while I could have afforded to finish my license, I realized owning my own airplane would have been prohibitive. Sure, I could get a Cessna 152 for around $20k at the time, but then hangar fees, maintenance, insurance, etc. were far more than I could afford.

There's a reason the average small plane owners at my airport drove Toyota rustbuckets with the bumper sticker "My other car is an airplane."

Are time shares not possible? Planes in this category are low utilization vehicles. The cost could be split between 3-4 or more owners.

Bottom line, between more pollution and making a hobby cheap , pollution should win every time.

> I personally look forward to the mainstream introduction of diesel engines into the US market that are able to run on Jet-A.

Seconded. 50 years ago diesel engines were too heavy for light aircraft, but the weight problem with diesels has been solved.

Wow, hearing that rich people are still allowed to dump lead into the atmosphere above our heads just ruined my morning. What things of non-entertainment value would we loose if we just banned leaded fuel overnight? Are crop dusters burning lead while spraying pesticides over our food crops?

You don't need to be rich to fly. You can buy a older used aircraft for the cost of a nice car. Sometimes cheaper. (Note: Not the Cirrus they're talking about upthread.)

They're working on replacing leaded fuel. The challenge is mostly regulatory, since they won't certify any of the replacements until they're sure it's 100% safe. It's still a work in progress.

As for what you'd lose: Search and rescue, medical transport (both commercial and volunteer), transport of blood samples for lab testing, high resolution aerial photography on your smartphone, powerline inspection to prevent forest fires, crop dusting. I could go on...

Search and rescue is almost soley helicopter traffic, and very rarely requires leaded fuel (a turbine doesn't need leaded, it can run on it, but it'll run just fine on unleaded too). The number of places still using non-turbine helicopters for anything other than training and very light sightseeing is very small, probably close to 0% worldwide.

Same is true of Medical Transport except possibly for Australia where light aircraft are used more frequently than turbine helicopters.

Same for powerline inspection, although there could be a couple of companies using Robinsons somewhere I guess.

Search and rescue in the USA is mostly done by the Civil Air Patrol, which almost solely uses fixed-wing aircraft.

Doing so would bankrupt pretty much the entire general aviation industry. Also, training new pilots would become impossible.

A lot of "low altitude only" pilots try to fly on MOGAS (car fuel, unleaded) when possible due to high cost of AVGAS 100LL (the most common aviation fuel, 100 octane low lead).

The problem comes with flying in lower atmospheric pressure (high or hot environment) because the replacement chemical used in place of the lead-based additives makes fuel much quicker to evaporate.

BTW, the really rich don't use AVGAS 100LL except maybe for hobby flying - kerosene-based JET A-1 is used in all turbine engines, as well as the rare diesel.

The amount of lead pollution produced by GA is a tiny fraction of the amount that used to be produced by cars, but still I think it's time for it to go away.

How do carburetors work when inverting the plane?

Most GA planes are only permitted to fly “positive g” for reasons like this...

Solutions to this problem were developed in the early 1940s: German fighters had fuel-injected engines.

https://en.wikipedia.org/wiki/Miss_Shilling%27s_orifice https://en.wikipedia.org/wiki/Pressure_carburetor

You could hold the director of the FAA at gunpoint and they still wouldn't let personal planes fly close enough to each other to make this remotely scalable. Even if you pulled the nonexistent vehicles out of a hat like magic we missed the social acceptance window. Automobiles emerged at a time when nobody really cared about safety regulations and they've continuously gotten safer, so nobody has a problem with the current 1.5 fatalities per 100 million vehicle-miles. These personal air vehicles could hit that value but it wouldn't be nearly good enough. We won't accept that in something new. We barely accept the airlines' current 0.2 deaths per ten billion passenger-miles safety record; they'll never accept a danger value 750 times as high.

> we missed the social acceptance window.

I don't think that's the reason we accept the higher death rate of cars. In cars, you feel like you have direct influence on whether you die, or at least you know the person in control. This might not be the actual case, but it's enough to make people save.

Compare this to planes: You're flown in big machines you don't understand by people you don't know and, if something goes wrong, you can do nothing but protect your head and hope for the best. Also, you have no way out - if an engine fails, prepare for quite some time of horror. Compared to a car, where you might have some unpleasant deceleration, but safety is always very near.

I'm pretty sure these factors are the main reason we are so wary of aviation deaths compared to car deaths.

Still, I think the age has something to do with it. The death rate for motorcycles is 24 times higher than that for cars but we would never allow a car missing enough safety features to be comparably dangerous on the road. You couldn't get regulators and the public to accept, say, 80 mile per hour electric rollerblades even if you could demonstrate they were just as safe as motorcycles.

> Still, I think the age has something to do with it.

No, it's the externalities. When you fly you don't just put yourself at risk, you put others at risk as well. Roads are designed to isolate faster-moving vehicles from people and property so the people who die are mostly the occupants of the vehicles and there is not much collateral damage. But it's not possible to isolate air traffic this way. So it's not just the occupants of aircraft who die, it's people sitting in their living rooms watching TV when a plane suddenly comes crashing through their roof.

Also, the limiting factor on air travel is not technology, it's takeoff and landing space. Even VTOL aircraft need a lot more room to takeoff and land than a car does because there's a blast zone.

Totally agree about the certification. It's so annoying having to do stuff like managing the mixture. There is absolutely no chance I'm going to be better at that than even the most basic FADEC system.

Every car since the early 90s has done away with the choke, but for some reason in a 2020-built C172 I still have to do this... :X

I would argue that this insistence on safety certification makes actual flights less safe. Because it results in it not being done, leaving something to the pilot who is more error-prone than an automatic system.

I haven't heard of accidents in GA aircraft that were due to poor mixture (though I haven't looked, and it could technically cause one if you set it too explosive) but when I still flew there were several incidents reported by our maintenance company who complained about cylinder scoring due to overheting.

It would be much better to have this automatically managed, and more environmentally friendly too, because there is no need for 'full rich' settings during takeoff/landing, it would just adjust it to ensure sufficient cooling. The full-rich is just a precaution to avoid the pilot miscalibrating it during this critical flight phase (and to avoid overheating on the ground of course).

Yes!

Rotax (Austrian company) actually builds great engines. They are heavily used in ultra-light airplanes with much more “relaxed” certification.

And you are right: in the ultra-light market, we see a giant boom because of lower certification hurdles and actually much safer systems: automated engine control, constant speed propellers, ballistic parachutes, etc. But Maximum Take-Off-Weight (MTOW) is 650 kg (Germany) that leaves you with a skinny wife/husband, some gas and light baggage...

Airplane engines are operating in tougher environments than car engines - hence bigger tolerances and less sophistication in some parts. They also must not fail. I mean, the engine has to be ok with starting up at 100 degrees F on the ground, climbing at maximum power to 10’000 ft, with below freezing temperatures within 15/20 min and descending down again with an urgy pilot pushing down the cylinder heat temperatures from 400 to 200 degrees fahrenheit within minutes (due to power reduction and increased speed cooling the engine).

Rotax engines are in fact fully certified aircraft engines as I understand it.

I am training for a PPL(G) in the UK (G is for Gyrocopter). There two proper certified makes of Gyro in the UK (Magni and AutoGyro), and all of the engines are Rotax. The newest is the 915, which includes a FADEC and turbo.

The fact that there are businesses that specialize in engine data monitor analysis and training on mixture operation says a ton about where GA is on this topic. There has been very little movement from the manufacturers on this.

That said highly recommend any materials from Mike Busch on this topic. Super knowledgeable and has moved the needle in getting people to understand how engines operate.

Yes, great recommendation!

I did hear about one accident due to mixture. They forgot to lean it out on a long flight and ran out of gas over a heavily treed area because they hadn't planned to run rich the whole way. If I recall correctly they survived.

Lycoming/clones with fuel injection, you typically only mess with mixture when starting and stoping them.

There is of course a minority religious sect within GA who follow lean-of-peak theory and mess with their mixture constantly for better mileage (and possibly better wear, but that can go either way).

I can recommend Mike Busch on that topic (like somebody did earlier).

Leaning is not only for better mileage. The engines usually do not like a fat mixture for prolonged periods of time... I can’t recall all the reasons, but I think a couple of mayor ones were: 1) lead in the avgas will foul the spark plugs for rich mixture settings which is a safety risk, 2) exhaust gas temperatures will be lower which can lead to a build-up at the exhaust valves which can lead to wear and failure and 3) you end up with more dirt in your oil which makes engine corrosion even a bigger possibility...

Lean if you love your engine. And read Mike Busch - because he has the facts better than I do...

Some argue that the US won the WW2 War in the Pacific by running LoP with the P-38 Lightning.

Charles Lindbergh demonstrated to US pilots how to do it, based on his Atlantic crossing testing, and the result was double the range. This allowed the US to surprise and "outrange" Japanese island installations, and to shoot down Admiral Yamamoto at great range.

(I would argue that the full story hasn't been told of the latter, since the planning was literally perfect. Something like radio navigation would have been needed, or incredible luck.)

Outrange was a Japanese strategy to attack US ships without expecting a return attack because of the Japanese planes having greater range initially. ie. LoP disrupted Japan.

AOPA: Technique: Lean-of-peak engine operations Keeping your cool on less fuel

https://www.aopa.org/news-and-media/all-news/2019/may/flight...

Wow, thats some cool fact I wasn’t aware of...

Aviation regulations are written in blood.

They are the only reason you can step into an aircraft with a reasonable expectation of making it safely to your destination. Disasters are generally not caused by obvious large issues, but a multitude of smaller compounding, seemingly benign causes that could otherwise be easily dismissed. The massive body of regulation around aviation is a direct result of this. Each one is almost always a direct response to a particular incident that killed people.

That regulations are motivated by avoiding deaths does not mean that they are effective at that, or even if they are effective that they are worth the costs. Regulations that specify specific safety equipment, rather than specifying target error rates or bonds/liability, are the worst because they get enacted and then stay on the books for decades unchanged.

They aren’t motivated by it. They’re a result of it. Practically every rule or standard in the book can be traced to a crash. Mode C veil? Aeromexico 498, 82 dead. Radio comms standardization? Tenerife, 583 dead. CRM: United 173, 10 dead. Altitude requirements while being vectored to an approach (91.175(i)): TWA 514, 92 dead.

I could go on.

The answer to “why can’t we just” in aviation is “because people died.” The FAA doesn’t make rules just to make them. They make them because aviation is unforgiving and it will kill you and your passengers if you screw it up.

It sounds like you're trying to disagree with the comment you replied to but none of those facts disagree with what they said.

Being caused by previous deaths and being motivated to avoid future deaths are very compatible.

> Each one is almost always a direct response to a particular incident that killed people

...in 1970.

Not saying that all aviation regulations are bad, but the fact that carburetors are still used in general aviation is pretty damning.

Interesting. Can you please elaborate your point about the carburetors? Why shouldn't they be used in general aviation?

I don't have much knowledge about engines, that's why I am asking you.

Carburetors are mainly susceptible to icing in normal use. Your engine can just start losing power and rapidly quit.

I once had carb icing in Hawaii at 2000' near Wheeler due mostly to humidity, and it was only thorough training that allowed me to figure that out fast enough to apply carb heat and save the plane.

A lot of inexperienced pilots just ride it down into the ground.

Dual fuel injectors prevent the icing problem (usually, unless your fuel actually crystallizes or freezes), but they have other failure modes like dirt or wear. Or forgetting to turn them on.

I'm not a pilot but I have a lot of experience with carburetors in cars and motorcycles.

Carburetors are fiddly analog devices that require tuning by way of changing port sizes, needle valve sizes, float levels, mixture adjusters, throttle position adjusters, and more. Most of these require disassembly, and they clog and distort as the carburetor ages. Carburetors are highly sensitive to temperature and altitude. Tuning them is "more art than science".

Modern fuel injection systems have digital control and feedback loops using a variety of sensors. They provide optimal fueling in pretty much any condition. They may be more complicated, but (at least in vehicles) they're rock solid reliable compared to carburetors. They also have more graceful failure modes - if the O2 sensor craps out, the computer defaults to running rich and puts a warning light on the dash.

I'm pretty sure that if the carburetor was invented today, it would never be allowed to carry human traffic.

Carbs have an advantage when you're not going inverted: You can gravity feed them, with no fuel pump required.It would be interesting to run the numbers on fuel pump engine outs vs carb icing engine outs to decide which one was worse.

A good example of a recent technical failure where one likely alternative is "everyone dies" https://en.wikipedia.org/wiki/British_Airways_Flight_38

The engine needs to be able to run without electrical power, or you need to convince the FAA that your backup will fail with an extremely small probability. This is one of the reasons magnetos + carburetors are common in general aviation

Nothing damning about it. It works.

I know seven people who have experienced engine failures in general aviation airplanes. Not one had anything to do with the carburetor.

Carburetor icing is pretty common cause of accidents, sometimes deadly.

However, that is often compounded with "the plane was made in 1960s"

Lol. And don’t forget the choke...

In planes it’s called the “primer” - and starting up the engines can be quiiiite tricky if you don’t know your plane. Especially in cold weather. I know plenty of people that ignited a carburetor fire on startup and the planes were almost write offs...

> how does anybody imagine that a novel aerial vehicle like this actually IS CERTIFIED within a lifetime?

All of the aircraft in existence have been certified in what is (for a very old person) one lifetime. I think we’ll be fine.

Their schedule seems to have 5 years for development, and 5 more for certification.

That seems reasonable.

> Their schedule seems to have 5 years for development, and 5 more for certification.

>That seems reasonable

That sounds laughable. Garmin, an established, well-funded, and well-connected company, spent multiple decades to certify one system with the FAA.

Seems like you lack of understanding about aviation in general.

You know why Piper/Cessna airplanes look exactly the same and keep using 50 year old engines? because of safety.

You know why pilot spends 30min pre-checking the aircraft? safety

You know why getting a pilot license is magnitudes more work? safety

Everything in aviation works around safety. If you want to innovate, sure, go ahead. If you want to innovate and do it safety and reliable, oops, that's going to cost you a lot and that's the same reason why there are not many players in aviation, engineering an aircraft or new powerplant is a big up-front cost with probably little return.

I'd argue that some of the difference between a pilot license and a driver's license would diminish if the difficulty of a driver's license more accurately reflected the challenges of operating a heavy vehicle in close proximity to pedestrians and other vehicles.

At the current standards, apparently the only hard requirement is a pulse.

Here are a few ways to kill you and all your passengers under instrument flight rules. Most of these involve a single mistake and a "controlled flight into terrain"

- Turn too early/late past an arbitrary point in space

- Turn the wrong direction

- Descend too early past an arbitrary point in space

- Use the wrong altimeter setting

- Failing to recognize and troubleshoot a failed instrument

- Failing to recognize signs of hypoxia

Not to mention dealing with emergency procedures while you don't do any of the above

Here is an example from a few years ago where a 777 missed a mountain by a few seconds after some non-standard instructions by ATC

https://www.youtube.com/watch?v=tFdXax7Zh_g

(I fly IFR as well)

I don't think the comparison between a regular driver's license and Instrument / IMC flight is reasonable. Perhaps more accurate to compare to a Commercial Driver's License with a HazMat load

Or driving a bus

Even in the aviation world, an IFR rating is largely considered the most difficult to get.

I would also add that in aviation, most accidents / incidents do _not_ result from a single error, but rather the stacking up of errors, which granted, are almost all variations of 'pilot error'

In a car, if in doubt you can just press and hold the brake pedal and you reach a safe condition. With the exception of motorways where you first have to pull over to the hard shoulder you can always do that.

In a plane you don't have that single goto option. You have to know the correct response to every possible situation and it's different in each case.

Plus, many mechanical failures in an aircraft will kill you. Most things in a car won't. You need to learn how to look after a plane, and what to check for.

You paint it to in a overly negative light, in my opinion. Much of the processes and procedures in place are not just functional, but behavioral training and conditioning to enable the pilot to troubleshoot mid-flight problems. It's like a highly capable human runtime exception debugger that will save the lives of everyone on board.

Some FAA regs and procedures are written in blood, but most are written in a way to prevent blood in the future.

I also like to echo the products Garmin is incorporating to increase the UX in the cockpit. They address some of the things you outline.

Remember, aviators aren't engineers, they are operators. The license requirement is there to ensure that pilots can do basic things like triage, malfunction diagnostic, ad hoc solution generation, energy management, communicate, etc.

For those interested, a couple of professors were using physics n hard math on the Lilium claims: https://www.aerokurier.de/8/3/7/8/5/9/6/aerokurier_Konzeptbe...

(It is in German though).

A big German magazine did a rather intensive “how much sense does it make” story: https://www.aerokurier.de/elektroflug/flugtaxis-kein-markt-f...

Safety, as measured in fatalities per person per mile per year, has increased by a factor of 100x since the 1960s: traffic increased 10-fold, while fatalities fell by the same factor.

So they do seem to be doing something right...

The lack of chaanges may just indicate that the current paradigm is a rather good one. If anvbody came up with a working scramjet, regulation would be a minor hassle.

It's really physics that are limiting here: supersonic travel is too inefficient both economically and environmentally. And personal autonomous local transport (i. e. "flying car") is impossible without some sort of breakthrough on noise.

If US and EU are so buerocratic that innovation is impossible, there will still be a big market for viable VTOL electric planes in the rest of the world.

Just like most 3rd world countries these days have better 4G coverage than supposedly 1st world countries like Germany, or rural US.

Check out: https://m.youtube.com/watch?v=_5K7cpxeXnc&feature=youtu.be

And guess the country :-)

I would never start such a business in the US or Europe.

Which third world countries have better 4G, than... say Estonia?

What 3rd world countries are you referring to? I'm pretty well traveled, and in my experience, this definitely has not been the case.

I suspect that aeroplanes will look very similarly in 100 years as well. There's such thing as physics that are the main reasons.

Don't look at fighter jets, as they are basically jet engines with guns and little control surfaces added.

As for airworthiness - there's a good reason for that - these things literally fly over your head.

Well, they should just use some of that VC money into pocketing (AKA lobbying) politicians to relax rules for these specific kind of airplanes, it has worked pretty well for the health industry in the US.

I mean, they probably do that.

But the point is: even if politicians DO push that and even if the rules and regulations are written down - doesn’t mean that they will be granted a certification.

I mean, funnily enough, they don’t yet even KNOW what they will have to adhere to and achieve and start burning through 300MM USD in investor money...

Boeing, Airbus, Cessna, Cirrus and Piper have dozens of years of experience and KNOW how to get planes certified - and yet they essentially “re-use older certifications”...

Fun times when they get told they need two ballistic parachutes or they need at least 30 min safety backup for hover flight (e.g., each Cessna pilot has to make sure he has 30+min fuel reserve). Your engineers will commit suicide when you surprise them with “oh there was this ask...”

And then the weather - I mean, seriously, there are super strict rules about visual minimas for flights in the US and Europe. If you go instrument rated things get even more complex and bureaucratic. Not to mention that instrument flight is PROHIBITED in Germany below 2000+ft in uncontrolled airspace...

Ah, and I want to see that flying electric lawn-mower make his way across the Alps with 50kts headwind, freezing temperatures at cruising altitude. Pilots landing in Innsbruck need special training for “Foehn” Approaches and when the heavy winds roll over the city and airport you can hear the pilots pushing full throttle in final to counter the 2000+ft/min downwash.

> e.g., each Cessna pilot has to make sure he has 30+min fuel reserve

Well, this is pretty much standard practice!

Edit: Ah I misunderstood your comment. So yes it is standard.

However I could see this not applying to these guys, because with vertical landing capability they should be able to support field landings with really minimal risk.

VTOL and several other major capability differences really should change the certification process. Current certifications a designed around certain engineering constraints.

Nearly all existing planes operate with the same basic parts, thanks to the fuel. This dictates the possible drivetrains.

Reduced complexity can lead to drastic reconsideration.

While I don't expect change to be super fast, regulation tends to move faster than normal when major disruption starts showing up in practice (drones, self-driving vehicles, ...)

There is already a certification process for helicopters and autogyros, so VTOL capability alone is no reason for a different certification process. Helicopters require a 20 minute fuel reserve in VFR conditions and 30 minutes in IFR conditions.

https://www.law.cornell.edu/cfr/text/14/91.151

Yes but these are automated, so they could be programmed to stay within X minutes of backup landing sites at all times, and altitudes to reach them easily.

No there will be a human pilot in control. It's going to be many years before fully automated aircraft are allowed to carry human civilian passengers through regular airspace.

Regarding the 30min safety: Yes, that is absolutely necessary... But I wonder how much 30min in safety will cost Lilium in actual range...

Helicopters in the U.S. only need a 20 minute reserve, at least for VFR.

Reply on news.ycombinator.com