Hacker News Reader

Comment by sohkamyung | Hacker News Reader

> Learning to deal with the intricacies of an ICE across the full range of flight conditions is half your training time wasted

I'm not familiar with flight terminology or flight training, so could you elaborate on the meaning of that sentence?

View on HN

ICE = Internal combustion engine

Off the top of my head:

- At certain temperatures and pressures you have to consider icing in the carburettor, which will block air flow into the engine.

- You have to tune (lean) the fuel input to reduce fuel consumption and prevent spark plug fouling, but not lean it too much that you overheat the engine.

Some more in-depth examples at http://philip.greenspun.com/flying/engine-management

And when descending, you have to be sure not to shock cool it etc.

Airplane gasoline engines are a mess, the technology is ~1950s at best and just hasn't advanced at all, with a few exceptions that, pardon the pun, never really took off.

A lot of that "mess" is due to a couple of critical requirements. Namely, low engine weight and reliability.

At this point it's mostly certification cost and liability insurance cost that's keeping the tech back. You can easily modify modern aircraft engines to be more reliable and efficient (PMAGs) but you need an STC (certification procedure) to make the switch so the experimental home builders are the only ones who actually benefit.

Rotax seem to be doing pretty well, they're used in a lot of popular trainers

Shock cooling as a concern for cylinder cracking is a fairly well debunked myth at this point, IMO, short of crashing your airplane into a cold lake.

Agreed the tech is old, but the uncommanded in-flight shutdown rate for pistons is also very low. Part of that is driven by high maintenance demands (some of which introduces failures themself), but a lot is due to the simplicity and redundancy.

Would I like variable spark timing (beyond impulse couplers for starting)? Would I like more automated starting, especially for hot starts? Would I like thermostatic cooling of cylinders? Yes, but not if they came with any substantial increase in failure modes. When I'm taking the family 700 miles at night in a single-engine airplane to grandma's for christmas, I'll suffer with arcane checklists and having to manually manage the engine if it reduces the chance of an uncommanded shutdown by even 5%.

> carburettor

Excuse me? Carburettor? In 2017?

> You have to tune (lean) the fuel input to reduce fuel consumption and prevent spark plug fouling, but not lean it too much that you overheat the engine.

Can't the engine management software take care of this?

Most flight schools have much older aircraft-- think 70s and 80s-- that have carbs. Newer aircraft are fuel injected, but often more expensive per hour.

VW Golf are fuel injected since 1975.

But they've only been able to fly since 2016.

Cars are different, though you could have pointed out that the Messerschmitt BF-109 had fuel injection in the 1930s.

An average car company makes millions of cars. Probably millions of a given model. Cessna has made ~44k 172 Skyhawks. It's one of the most produced aircraft, and it's been in production largely unchanged since 1956. I think Tesla is going to make more Model 3s this year than Cessna is going to make Skyhawks.

Don't they cost hundreds of thousands of dollars? Surely you're going to find an engine manufacturer that can build you an engine that's not half a century old.

If it ain't broke...but the 172[R and S] has been using a fuel-injected engine since 1996[0]

[0]https://en.wikipedia.org/wiki/Lycoming_O-360

And the flight hour cost of one of these is double the price of a 1970s/80s carbureted 152. Guess which one the majority of student pilots are learning on.

Of course, and if you want something like that, you can find aircraft manufacturers using more modern engines. Want a car engine in an airplane? You can have that. Want a fuel-injected digitally-controlled turbocharged water-cooled diesel? Step right this way.

The big, old manufacturers like Cessna tend to be really conservative, though. They have something that's Good Enough and they have little incentive to make radical changes. Part of this is because a light aircraft's powertrain is safety-critical, unlike automobiles. In a car, if your engine explodes, you're stranded. In a small airplane, depending on where you are when it happens, you're probably hurt or killed. If you make a radical change to your engine, and that change results in a failure that kills a customer, it looks bad. Newer manufacturers seem more willing to push the envelope in this respect.

What's this "engine management software" you speak of? :-)

Seriously, GA aircraft engines are seriously old tech.

You can indeed get engine management computers, but the thinking in the aviation world is that this would be just another thing that could go wrong. In any case, if the computer were to fail, the pilot would have to know how to manage the engine manually. Aircraft engines have various design choices designed to make the things more reliable, such as dual independent ignition systems that are isolated from the battery. The pilot deliberately has independent control over things like fuel/air mixture for good reasons.

ICE = Internal Combustion Engine.

Reply on news.ycombinator.com