Live data from Hacker News

Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

news.ycombinator.com

211–220 of 632 posts

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#211
post #72

Earlier quoted context omitted.

Take a look at the more economical modern GA aircraft. A Cirrus SR22 can run at 10 gallons per hour, with a ground speed of 160+ mph. Yes, it’s more total fuel consumption than a car (but in an hour covering 2x the distance, and allowing to travel more directly) but not at all close to turbine or turboprops. At the extremely cheap (accessible to more pilots) side for pressurized planes, fuel burn is going to be 40gph…

yep! lots of interesting stuff indeed! Our plane will be ~7 gph at cruise burning unleaded fuels and fly ~170mph over the ground (with no wind)

For my part, I find it questionable to invest in and develop a transport technology that consumes seven times as much fossil fuel as a car.

You give consumption of 7 gallons in cruise mode. I don't want to know what is burnt during take-off or landing.

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#212

I was shocked when learning to fly how much GA tech was stuck in some weird pre-World War II era state. Everything seems so needlessly complicated. Even small aircraft cost so much that I couldn’t imagine a computer controlled system and fly by wire would be cost prohibitive. Even a new C172 is close to $400k at the low end, and anything you’d actually want to fly is much more. A few grand in electronics seems worth…

> It feels like the real problem in this industry is lack of competition. There’s a chicken and egg problem: planes are expensive because so few people buy them, but so few people buy them because they’re expensive. Lack of competition is not the root cause. It's all the regulations. Certifying a new GA airplane takes a lot of time, and costs a fortune. This translates to high development costs, and certification cos…

One key difference I see between something like this and an ultralight is the danger to people on the ground. A few thousand pounds moving at up to 200mph or so (potentially 300+ after structural failure) is a lot of kinetic energy.

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#214
I am sometimes flabbergasted by some of the things YC will fund. This is one of them.

I don't know where to start, other than to say I would never make some of these assertions in public. Doesn't YC provide legal support to the startups it funds?

For example:

"Unlike a traditional airplane, it becomes impossible to command the airplane into a stall, a spin, unsafe attitudes, or other bad states."

This is categorically and absolutely false and wrong. There is no such thing as "impossible". Do not ever say things like that or you will find your own words used as weapons against you in the inevitable court case when something does go wrong. Too late now, it's already out there.

Attorney: Sir, your plane stalled and became uncontrollable. The pilot had no controls because you removed pedals and any industry-proven direct control mechanisms. Your company asserted this failure mode was impossible, yet, it happened. Exhibit A, in your own words.

"we can use MEMS gyroscopes that cost Attorney: Sir, your aircraft killed all four of its passengers, destroyed three homes, killed an additional three people on the ground --two of them children-- and sent half a dozen others to the hospital in critical condition. Why did you use cheap sensors instead of industry-standard, battle-tested, fully-vetted, qualified, susceptibility-tested hardware?

"We’ve developed our own control surface actuators"

Attorney (following-up to the previous question): Control surfaces failed to actuate controls as required. You used cheap self-designed actuation hardware that did not have extensive engineering design, testing and qualification pedigree as industry-standard products...to save money...

"own radios and GPS hardware (an aviation grade GPS can cost upwards of $10k, but it’s the same hardware as in a $20 consumer grade GPS)"

Attorney: I repeat the question. Your cheap $20 GPS just killed half a dozen people. How could you justify placing such little value on human life?

"The only real single point of failure is the engine."

False! A triple-redundant design, will, at best, mitigate a single failure and ZERO in the case of common-mode failures. Triple modular redundancy requires, at a minimum: Triple redundant power, triple sensing, triple compute, triple actuation and no common-mode failure mechanisms. Once a triple-redundant system experiences a single failure, it turns into a system with two possible sources of truth, which means it is impossible to know who might be right.

My free advice is to stop talking and get out of this business before you kill people. Seriously. You are not at SpaceX with hundred-billion-dollar budgets launching craft that fly so far away from population centers that they could explode and nobody gets hurt. Before Crew Dragon flew for the first time billions of dollars went into evolving the systems that made it possible.

Then again, what do I know? I could be wrong.

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#215
...three flight controllers ... two different sets of sensors ... independently compute the desired actions to take, and vote on what to do. Each ... has its own power source, backup battery, networking hardware, and set of critical sensors.

You didn't mention activators. Do all the redundant components you mentioned connect to a single voting system, and single set of actuators? How do you avoid your voting system and actuators being another single point of failure?

Or does each redundant system have it's own set of actuators, and enough of them to tolerate failure of one set? (e.g. three sets sized such that two working ones can overpower one erroneous one)

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#216
post #83

Instrument-rated pilot (and engineer) here. First - congrats on the launch! I think you're working on an interesting set of components that will prove useful to GA aircraft technology. Bringing fly-by-wire, and lowering the cost of maintenance/manufacturing are both great efforts. That being said, my personal view is that stick-and-rudder control is one of the less critical components to improving GA safety. Everythi…

Excellent points, it's really these other aspects that get pilots into trouble. For example it's crazy that most pilots are still taught to calculate W&B using printed charts and approximate takeoff performance. I think you could save more general aviation lives with a fairly minimal system. A gas gauge sensor that calculates whether you have enough fuel to get to your destination + reserve. Avionics where you input…

As much as I am excited about Airhart, I would be way more excited for all of these items than I would be for the fly by wire bit. These would be more likely keep me and my wallet in one piece and lower the cost of aviation. Unfortunately they're not terrifically sexy or super lucrative problems.

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#217
post #206

Earlier quoted context omitted.

> For example it's crazy that most pilots are still taught to calculate W&B using printed charts and approximate takeoff performance. We're taught that so we know what we're doing. Then we open our iPhone app (EFB or electronic flightbag) and do it there. In fact, part of the reason we do it there is because most of them phone home to their maker and log that we did it. So if there's an accident people can know "well…

Stall speed is a misnomer. Stalls occur due to exceeding the critical angle of attack and can happen at ANY speed.

I think that was his point. To calculate if you are going to stall in a final, you need to know your weight and flap setting to see if you will exceed the critical angle of attack when doing the turn, exactly because the speed isn’t constant.

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#218
post #210

Earlier quoted context omitted.

The barrier to entry is not learning how to fly. That is the easiest thing for most people.

One of the biggest barriers at this point is probably getting a medical. There are tons of perverse incentives there - getting one if you've ever been prescribed mental health meds, for instance, can be at best a ton of red tape.

I’m willing to bet the major barriers are money and time.

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#219

I was shocked when learning to fly how much GA tech was stuck in some weird pre-World War II era state. Everything seems so needlessly complicated. Even small aircraft cost so much that I couldn’t imagine a computer controlled system and fly by wire would be cost prohibitive. Even a new C172 is close to $400k at the low end, and anything you’d actually want to fly is much more. A few grand in electronics seems worth…

> It feels like the real problem in this industry is lack of competition. There’s a chicken and egg problem: planes are expensive because so few people buy them, but so few people buy them because they’re expensive. Lack of competition is not the root cause. It's all the regulations. Certifying a new GA airplane takes a lot of time, and costs a fortune. This translates to high development costs, and certification cos…

I do believe that’s a factor (and mentioned it further down) but there are shockingly few companies that build planes.

The regulatory cost is amortized over the number of planes built. If people bought planes the way they buy cars, they’d cost far less. Hence the chicken/egg problem I mentioned.

Re: Launch HN: Airhart Aeronautics (YC S22) – A modern personal airplane

#220

> In the US, trips that are 50-300 miles are almost all done by car because that distance is too short for commercial airlines and too far for public transportation. A bit off-topic but this is a political problem, not a technical one. Trips of 50-300 miles are certainly within the operating range of fast and efficient rail travel as demonstrated in multiple places around the world outside of the US.

Interestingly I'm hosting my dad near Amsterdam for the week and to get from Sacramento to SFO (from where he flew to AMS) he... flew. Which I guess is logical, but seems insane to me when he lives close to Sacramento Amtrak which actually HAS good, regular service to SFO (via Bart, connecting at Richmond). Even if you build the train some people won't give it a try, sadly. A lot of carbon was spewed in to the air to…

I saw a cargo flight going from LAX to Ontario, California which looks like ~80km/50mi distance, which seems crazy to me.

I don't live in Europe, but I was there earlier this year going from Amsterdam to Paris. That's 430km/260mi direct from what I can find. I took the train, it took 3.5 hours, direct from central Amsterdam to central Paris, and no security theatre. And I had leg room!

Post reply on HN