The glaring weaknesses in flight training

[Rich]

Recent safety presentations from FAA, EAA, AOPA, etc. have started focusing on the "startle" factor. Usually this is presented in the context of engine failure, especially on takeoff. But this is also a factor in other conditions as well. The greater problem is we don't teach for the unexpected case. Think about it:

1. Practice for engine out? Reach up and pull the throttle back (never dare on takeoff unless you're nuts). Student sees this and is forewarned.
2. "We're gong to do stalls now." Student gets sweaty palms but knows what's expected. Worse we slow down and pull the nose above the horizon, imprinting false understanding on what initiates a stall.
3. "We're going to practice spins now." Student gets really sweaty palms but same as the stall.
4. "Let's do some crosswind landings." This isn't helpful when some other day where a sudden gust appears out of nowhere or the wind is shifting all around.

So what happens when one of these events happens without the pilot intending to do so - especially when you're all tensed up close to the ground? Startle/freeze and maybe crash/die.

I can think of three cases when I was a CFI where I saw the guy in the other seat become a passenger with no idea what to do. One I induced intentionally (surreptitiously pulled the mixture). The others were of their own doing and I had to take over.

No real solid ideas how to correct this problem. For example in years past spin recovery was part of private pilot training. There are those that say bringing it back would cure the stall/spin problem. But it was discontinued because it didn't seem to result in a reduction of inadvertent stall-spin accidents and there were lots of accidents resulting from performing spins for training purposes. And many aircraft these days are not suited for intentional spin training.

[dant]

I spend a lot of time mulling over why the time-to-upset is so low for VMC in to IMC incidents - especially with modern avionics. Given that we can't ever ask the pilots why, all we can really do is pontificate, and my suspicion is it's this exact thing.

I personally would like to see flight simulator improvements. It's a lot easier to surprise someone when they can't see you prepare for it. The problem is the simulators we have access to are basically useless for flight characteristics and are more for learning and managing avionics. The best we can do is an instrument failure which doesn't seem to be what's killing folks. Get something where you can actually have a reasonable sense of control of the aircraft and do engine out / windshear training then.

You probably won't get the same feel of your stomach falling out from under you when you suddenly drop due to windshear but it's better than nothing, and make with some advances in motion tech they can model that reasonably enough.

[Rich]

I agree about simulators - at least the ones available for our light GA planes. They're laughably incompetent at emulating flight characteristics.

But if full-screen simulators could be made available that faithfully emulated flight characteristics perhaps these might be a big help. You could have an evil instructor/mentor/sadistic nun or whoever behind you initiate spins/stalls/weird winds without warning.

[mhoran]

Early on in my flying career I made all the mistakes during my first VFR into IMC encounter. I had only about 90 hours at the time, and was embarking on my first real cross country on my own (outside of training) with a passenger (who got sick early on in the flight yet we continued on.) There was a presidential TFR over my destination so I was unnecessarily nervous. On an otherwise beautiful VFR day, there was a bit of cloud just between where I was and where I was going. It was at my altitude (~4500 feet) and I could have gone around it, above it, or below it. But I misjudged the cloud location, and flew right into it. I freaked out. I didn't want to deviate from course, or tell ATC what I'd done, because of the TFR. I did all the wrong things. I started fighting with the trim, but ended up nearly stalling the plane. My altitude and airspeed was all over the place. I could have stall spun right out of the cloud. Fortunately, I made it through without issue. If I'd just kept flying the plane, all would have been fine. I just wasn't thinking rationally and became a passenger, as Rich suggested.

I've subsequently gotten my instrument rating, so VFR into IMC really isn't much of an issue these days. But a little while back I did encounter some unforecast IMC along my route of flight and had to turn around and request a clearance to proceed. This time things played out much differently. I already had flight following (as before), so it was very easy to request pop-up IFR, even from New York approach. I now had a plane with full autopilot, which I was comfortable using. I was able to dial in a 180 degree turn and did not penetrate the clouds. I was able to command a controlled descent and avoid VFR into IMC. Before I'd even completed my turn I had my clearance and was back on course, and it was very much a non-issue.

The difference between the two scenarios is experience. It's a bit silly to say, but one takeaway is that in order to avoid VFR into IMC I had to get an instrument rating. I suppose the 800 or so hours I've picked up in the subsequent years also helped. But it's super hard to drill this all into students heads through training. Most of the time it just goes in one year and out the other, given the task saturation we're under while flying. Unfortunately many of these incidents happen early on in a pilot's career, as was my experience early on. Even though I knew *not* to fly into cloud, I did it anyway. Even though I had ATC available to help, I didn't use them. I certainly wouldn't do that now, but with just a few hundred hours, it's very easy to make these critical mistakes.

[dmloftus]

I think it is even more scary when you consider new MOSAIC proposals potentially allowing sport pilots to fly larger, more powerful aircraft. We all want to increase GA, but sacrificing safety to do so is crazy. I think the PPL PTS should include more basic instrument instruction, preferably in IMC, to get student pilots more comfortable in getting out of inadvertent VMC into IMC. And flight schools should price in the first 10-20 hours of instrument training in their PPL training packages. Like the earlier post, the statistics on how quickly an inexperienced pilot goes down in IMC is very surprising.

[Steve]

So what happens when one of these events happens without the pilot intending to do so - especially when you're all tensed up close to the ground? Startle/freeze and maybe crash/die.

As if on cue for this discussion:

https://www.youtube.com/watch?v=d5yt3lCQtSM

Sad to watch...

[Rich]

So what happens when one of these events happens without the pilot intending to do so - especially when you're all tensed up close to the ground? Startle/freeze and maybe crash/die.

As if on cue for this discussion:

https://www.youtube.com/watch?v=d5yt3lCQtSM

Sad to watch...

Actually it was the cue for this discussion.

[michael.g.miller]

I personally would like to see flight simulator improvements. It's a lot easier to surprise someone when they can't see you prepare for it. The problem is the simulators we have access to are basically useless for flight characteristics and are more for learning and managing avionics. The best we can do is an instrument failure which doesn't seem to be what's killing folks. Get something where you can actually have a reasonable sense of control of the aircraft and do engine out / windshear training then.

I agree wholeheartedly. GA is unbelievably lacking with regard to flight simulators. And our safety record leaves more than a little room for improvement.

When I did my ATP CTP training, I was blown away by the training the pros do. I did about 5 hours in a full motion sim (both a Global and a Falcon). It was night and day compared to GA. The sim instructors were pros, tripping you up on tough approaches, failing instruments and equipment at the worst time. Exactly what you want in a sim session. We were able to quickly iterate through scenarios with the click of a mouse. It was a workout!

I left the ATP CTP training thinking... why can't GA be like this? Then I remembered. The course was around $5k. Can you imagine GA pilots shelling out $5k every 6 months for recurrent training? Everyone would throw their hands up. And we're left with our current accident rate.

[dant]

Can you imagine GA pilots shelling out $5k every 6 months for recurrent training

Well, the real question is whether the insurance companies can imagine requiring it for decent rates, and whether the claims/accidents they see would be affected by this training.

I suspect they'd rather see 5k every 6 months shelled out on fuel calculation training.

[michael.g.miller]

I don't think insurance companies would mandate this kind of training. To put it in perspective, most people pay $5-10k for the whole year of insurance. The insurance companies would either have to:

- Discount the insurance to be practically free
or
- Effectively double your insurance premiums

By the fact that insurance is the same cost (or less) than recurrent full motion sim training, the market has spoken. Sim training might reduce accident rates, but is not economically worth it. At least to the underwriters. Of course if claims get a lot more expensive, then this calculus might change.