GA Crash at PDK

[Ed Haywood]

Flaps add drag and reduce glide distance. So turn clean until field environment is made, then add flaps to slow touchdown speed and shorten rollout. At least that's how I practice it, or would if I had flaps.

Question for the crowd: full flaps if you elect to land straight ahead?

 

[kyleb]

You need to do a 90+270 turn to get back to the runway.

This is the part people get hung up on. You DON’T need to make it back to the runway. Just to somewhere relatively flat with no major obstacles. The dirt or grass between the taxiways (or any similar place) would be fine.

And I’d check your turn geometry.

 

[write-stuff]

 

[write-stuff]

Look online for videos. A lot depends on the angle of climb of the aircraft and glide angle. Some aircraft cannot get back to the runway. You need to work this out at altitude.

Yep, it's all about the angles. But most of the YT videos about it are crap.

 

[Flocker]

Crap, it was N103AV, I know that airplane.

I've flown in that plane. Kudos to the PIC. CAA is an excellent flight school.

 

[sarangan]

The pilot did a great job. He flew it into the ground instead of trying to arrest the sink rate and stalling. The impact looks bad on the video only because of the high forward speed. The outcome speaks for itself. Only minor injuries. I think there is something very useful to learn here.

 

[Half Fast]

Sounds like the best option is to pick a good crash site as part of the pre-flight...

Pick it as part of the flight plan and brief it as part of the pre-flight.

Google Earth can be a big help during flight planning. Look over the area and know your options for each runway. And if the engine doesn't fail, take a good look at the area and figure out where you'll go if it fails on a future flight.

Unfortunately, for a couple of the runways at Winter Haven a lake is probably the best option. Ditchings are usually survivable which is what matters, but it's guaranteed the plane will be a total loss. But a good landing is any landing you can walk (or swim!) away from.

 

[Racerx]

Pick it as part of the flight plan and brief it as part of the pre-flight.

Google Earth can be a big help during flight planning. Look over the area and know your options for each runway. And if the engine doesn't fail, take a good look at the area and figure out where you'll go if it fails on a future flight.

Unfortunately, for couple of the runways at Winter Haven a lake is probably the best option. Ditchings are usually survivable which is what matters, but it's guaranteed the plane will be a total loss. But a good landing is any landing you can walk (or swim!) away from.

I'll take a lake over a subdivision any day

 

[Pinecone]

The 90-270 builds in a bit of a safety factor due to the surprise of the situation.

And your diagram requires some time after your 210 degree turn due to the offset. So it really isn't much faster, if at all.

 

[Pinecone]

Flaps add drag and reduce glide distance.

Question for the crowd: full flaps if you elect to land straight ahead?

Try it both ways. The first part of the flap extension increases lift more than it increases drag. So in many cases a slight flap setting can increase glide distance.

That is why some aircraft have take off flap setting.

 

[flyingron]

I lost an engine on takeoff flying young eagles a while back. Straight ahead wasn't an option. I figured I could put it down on the airport access road but as I got to into the turn I knew I could make the (large empty) ramp or the taxiway. As I was finally lined up I'm thinking, "what am I forgetting." Oh yeah, landing gear. Fortunately, it comes down fast in the Navion. I had a little bit of a side motion (which rolled the main off the rim) but otherwise the plane wasn't hurt much.

 

[hindsight2020]

Um, nope. Flaps that do not increase the wing area of the planform, do not contribute to an extension in glide distance. There are flaps that do, the Piper -28 series is not one of them. L/D discussion is conflated here.

Also, takeoff flap settings are not established for glide extension benefits. They're established for climb gradient, obstacle clearance, liftoff speeds and therefore refusal speeds, and runway available considerations. None of which have anything to do with engine out flying characteristics.

 

[kyleb]

Um, nope. Flaps that do not increase the wing area of the planform, do not contribute to an extension in glide distance. There are flaps that do, the Piper -28 series is not one of them. L/D discussion is conflated here.

But, if you were in a situation where your airspeed was bleeding down, there was no way to recover it, and you needed the wing to continue to fly for a couple of seconds longer than it was gonna fly in the slick configuration, a notch or two of flaps might buy you that.

 

[dtuuri]

Flaps that do not increase the wing area of the planform, do not contribute to an extension in glide distance. There are flaps that do...

I suppose anything is possible, but how would extending the flaps improve L/D max? Seems it would increase drag to me, else why not fly with them extended all the time?

Also, takeoff flap settings are not established for glide extension benefits. They're established for climb gradient, obstacle clearance, liftoff speeds and therefore refusal speeds, and runway available considerations. None of which have anything to do with engine out flying characteristics.

Allow me to pick a nit for the student pilots: As to the part about "climb gradient", flaps only reduce ground run. Steeper airborne climb gradients require less flaps, a daily operational fact of life out West for air carriers.

 

[dtuuri]

But, if you were in a situation where your airspeed was bleeding down, there was no way to recover it, and you needed the wing to continue to fly for a couple of seconds longer than it was gonna fly in the slick configuration, a notch or two of flaps might buy you that.

You mean like Ernest Gann did to keep from crashing into the Taj Mahal in Fate is the Hunter?

 

[brcase]

I'll lay odds that the CFI flys gliders

At a minimum regular practiced that emergency, as any good CFI should. For that matter any good pilot.
Deciding if a turn back is possible should be an individual choice based on experience, the Aircraft and location.

For me there are quite a few situations where I wouldn't. However there are many where I have a very well defined point as to when It changes from the impossible to the possible turn.

Comes down to that item on many check lists “what is your abort Plan?” This should cover everything at least up to pattern altitude.

Brian
CFIIG/ASEL

 

[Tools]

I have even begun to rethink the “don’t turn till 500 feet”.

If your engine quits anywhere along that path, all that beautiful runway DIRECTLY BENEATH you is the one place on earth you can’t possibly get to…

Conversely, if you immediately turn 30 deg towards the worst terrain around, it now puts where you DONT want to go directly beneath you, leaving nothing but better options.

 

[nauga]

Um, nope. Flaps that do not increase the wing area of the planform, do not contribute to an extension in glide distance.

Flaps that increase CLmax, as they all tend to do, give you the option of flaring a little bit more or a little bit longer, while bleeding off additional airspeed. Steady state, no; but at the endgame, yes.

Nauga,
unsteady

 

[flyingron]

But, if you were in a situation where your airspeed was bleeding down, there was no way to recover it, and you needed the wing to continue to fly for a couple of seconds longer than it was gonna fly in the slick configuration, a notch or two of flaps might buy you that.

But it won't. While you could fly at a lower speed, the increased drag is going to slow you down faster.

 

[kyleb]

But it won't. While you could fly at a lower speed, the increased drag is going to slow you down faster.

Dude...

 

[Fearless Tower]

Dude...

Sounds like he’s never come up short on a power off landing in an Arrow with Johnson Bar flaps….

 

[Ed Haywood]

Flaps that increase CLmax, as they all tend to do, give you the option of flaring a little bit more or a little bit longer, while bleeding off additional airspeed. Steady state, no; but at the endgame, yes.

Nauga,
unsteady

Which is why I asked if full flaps would be preferred in an off airport landing. Assume glide distance not a factor and you have to put it into a residential area, parking lot, woods, or other unsuitable area. It's going to be a crash, just a question of how bad. Is it generally better to put in full flaps before impact to reduce airspeed and energy?

 

[kyleb]

Which is why I asked if full flaps would be preferred in an off airport landing. Assume glide distance not a factor and you have to put it into a residential area, parking lot, woods, or other unsuitable area. It's going to be a crash, just a question of how bad. Is it generally better to put in full flaps before impact to reduce airspeed and energy?

Absolutely. Even a few knots less speed greatly reduces the kinetic energy you need to dissipate in the arrival.

 

[Pinecone]

Um, nope. Flaps that do not increase the wing area of the planform, do not contribute to an extension in glide distance. There are flaps that do, the Piper -28 series is not one of them. L/D discussion is conflated here.

Also, takeoff flap settings are not established for glide extension benefits. They're established for climb gradient, obstacle clearance, liftoff speeds and therefore refusal speeds, and runway available considerations. None of which have anything to do with engine out flying characteristics.

Which is why I said to TEST in your airplane.

 

[dtuuri]

Which is why I said to TEST in your airplane.

It's fun to test, but you'll only find that this conclusion is most likely incorrect, intuitive though it may be:

Try it both ways. The first part of the flap extension increases lift more than it increases drag. So in many cases a slight flap setting can increase glide distance.

You're forgetting that you correctly said "...more than it increases drag." So, you're admitting drag also increases and therein lies the rub: When drag is increased glide distance is decreased. (You need to point the nose down more to get more thrust from gravity to offset the increased drag.)

 

[brcase]

Flaps add drag and reduce glide distance. So turn clean until field environment is made, then add flaps to slow touchdown speed and shorten rollout. At least that's how I practice it, or would if I had flaps.

Question for the crowd: full flaps if you elect to land straight ahead?

Practicing Power of 180’s will really help you understand how the flaps can help.

I agree clean until the runway/landing environment is made. If still high use the flaps and/or a slip to get down if needed. Usually adding flaps to extend distance to the touch down point, really only works once you are close to or in Ground Effect.

As for the straight ahead landing question if there is any doubt about getting stopped, immediately going to full flaps while still airborne will usually get you stopped quicker.


Brian
CFIIG/ASEL

 

[write-stuff]

The 90-270 builds in a bit of a safety factor due to the surprise of the situation.

And your diagram requires some time after your 210 degree turn due to the offset. So it really isn't much faster, if at all.

With all due respect, I completely disagree. 90-270? How exactly does that work? I don't see any safety factor in it at all. It is certainly longer periods in the turn(s) and probably takes you further away from the airport. Could you draw this out?

 

[nauga]

...therein lies the rub: When drag is increased glide distance is decreased. (You need to point the nose down more to get more thrust from gravity to offset the increased drag.)

Not in the short term, as a normal flare demonstrates. Imagine you're just about to touch down, having glided all the way down at L/Dmax for best glide range. Flaps up, you can increase angle of attack, thus increasing drag but also increasing lift to (temporarily) trade kinetic energy for potential and shallow the descent while bleeding airspeed. Now very close to stall AOA, if the flaps are lowered (some) lift and drag both increase again, but you've got a little more lift margin so by maintaining (or lowering, even) that AOA you can glide even further if you manage it well. In the long term, if you had the altitude, you are correct that the overall glide would be shorter than if you held L/Dmax the whole way, but if you touch down before you have to give back that KE for PE trade you made you'll be farther. Maybe not by much, but still...

Nauga,
who preaches what he parctices

 

[dtuuri]

In the long term, if you had the altitude, you are correct that the overall glide would be shorter than if you held L/Dmax the whole way,

Yes, I'm referring to long term, since that was how I understood Pincone.

but if you touch down before you have to give back that KE for PE trade you made you'll be farther. Maybe not by much, but still...

You've lost me here, though. Here's another way to say it, maybe?
During WW2 an American pilot almost crashed into the Taj Mahal – Historical articles and illustrationsHistorical articles and illustrations | Look and Learn

 

[nauga]

Yes, I'm referring to long term, since that was how I understood Pincone.

Haha...I was just about to comment on his post as well.

Nauga,
onward and upward.

 

[nauga]

Try it both ways. The first part of the flap extension increases lift more than it increases drag. So in many cases a slight flap setting can increase glide distance.

At steady-state glide distance is a function of L/D, with max range at L/Dmax. If you've found a flap and setting that increases L/Dmax one would wonder why they didn't just shape the airfoil that way as a baseline. :-/ Do you also get better range with the flaps extended?

I assume you're not talking about reflex.

Nauga,
straight-aero

 

[nauga]

Which is why I asked if full flaps would be preferred in an off airport landing. Assume glide distance not a factor and you have to put it into a residential area, parking lot, woods, or other unsuitable area. It's going to be a crash, just a question of how bad. Is it generally better to put in full flaps before impact to reduce airspeed and energy?

Generally, anything that reduces kinetic energy is better. The timing of the application of full (or *any* for that matter) flaps can make it better or worse. If you think you're going to use a tool, practice practice practice.

Nauga,
and the DIT ('dog in truck') approach

 

[RyanShort1]

One of these patterns was not like the others. Did a practice turn from about 300’ agl at Llano in a Luscombe today. Even a perfectly expected and executed turn is a serious pucker factor.

 

[FastEddieB]

Generally, anything that reduces kinetic energy is better.

A while back I came up with an Excel spreadsheet to show the increase in energy that results from an increased landing speed. I looked up the stall speeds of a C172 and came up with a typical 53 kts clean and 48 kts w/full flaps. Instead of my spreadsheet I thought I’d try ChatGPT:

“What is the percentage increase in energy of an airplane landing at 53 knots compared to one landing at 48 knots?”

The kinetic energy of an airplane is proportional to the square of its velocity. Therefore, the percentage increase in energy of an airplane landing at 53 knots compared to one landing at 48 knots can be calculated as follows:

First, we need to calculate the ratio of the kinetic energy of the two airplanes. Let KE_53 be the kinetic energy of the airplane landing at 53 knots, and KE_48 be the kinetic energy of the airplane landing at 48 knots. Then:

KE_53/KE_48 = (v_53^2)/(v_48^2) where v_53 and v_48 are the velocities of the two airplanes.

Substituting the given values of v_53 = 53 knots and v_48 = 48 knots, we get:

KE_53/KE_48 = (53^2)/(48^2) = 1.2207

This means that the airplane landing at 53 knots has 1.2207 times the kinetic energy of the airplane landing at 48 knots.

To calculate the percentage increase in energy, we can subtract 1 from the ratio and then multiply by 100:

Percentage increase in energy = (KE_53/KE_48 - 1) x 100% = (1.2207 - 1) x 100% = 22.07%

Therefore, the airplane landing at 53 knots has approximately 22.07% more energy than the airplane landing at 48 knots.

(bolded mine)

22% more energy can mean the difference between no injury and minor injury, between minor and major injury, or even major injury and death. So, yes, full flaps and landing at or very near stall speed can be critical in an off airport landing.

As an aside, who needs a spreadsheet for this sort of query any more?

 

[Half Fast]

Substituting the given values of v_53 = 53 knots and v_48 = 48 knots, we get:

KE_53/KE_48 = (53^2)/(48^2) = 1.2207

Hmmm......

When I punch 53^2 / 48^2 into my trusty HP-15C I get 1.2192, not 1.2207. That's only a 0.12% error, but still it makes me wonder about ChatGPT's arithmetic abilities. How does a computer make this sort of error?

Maybe you shouldn't be too quick to abandon your spreadsheet, and maybe we shouldn't trust ChatGPT too much until it's had lots and lots of verification testing.

 

[FastEddieB]

Interesting. I think my spreadsheet was set up to yield percentage in nearest round number, i.e. 22% in this case. But that was just the way I formatted that cell. I’ll check tomorrow to see if my spreadsheets agrees with your calculator when taken out to the same number of digits. I suspect it will.

Much has been made of ChatGPT sometimes giving wrong or inaccurate answers. Hopefully those errors will diminish over time.

edited to add: Wolfram Alpha returns 1.21918402777… So looks like you’re right.

 

[4RNB]

What is the best way to train/prepare for (what I'm assuming to be) an engine failure below 500' AGL on takeoff? Legitimate question.

sim?

 

[FastEddieB]

I know this is a bit of a derail, but I still find it interesting, so forgive me.

“It appears your calculation of KE_53/KE_48 = (53^2)/(48^2) = 1.2207 is imprecise. Any idea why that may be so?”

The answer, in part, still appears to be wrong:


You're correct, my calculation for KE_53/KE_48 = (53^2)/(48^2) = 1.2207 is imprecise. The actual value for this calculation is 1.2653.

The reason for this discrepancy is that I made an error in my initial calculation. Instead of calculating (53^2)/(48^2), I accidentally calculated (53/48)^2. This gave me a value of 1.2207, which is incorrect.

To correct this, the correct formula for the ratio of kinetic energies should be:

KE_53/KE_48 = (53^2)/(48^2)

Using this formula, we get:

KE_53/KE_48 = (53^2)/(48^2) = 1.2653


Weird how in its attempt to give human-sounding answers it appears to make human-like mistakes!

 

[Pinecone]

It's fun to test, but you'll only find that this conclusion is most likely incorrect, intuitive though it may be:

​

You're forgetting that you correctly said "...more than it increases drag." So, you're admitting drag also increases and therein lies the rub: When drag is increased glide distance is decreased. (You need to point the nose down more to get more thrust from gravity to offset the increased drag.)

Glide distance is based on LIFT divided by drag. If you increase the lift more than you increase the drag, you get a longer glide.

 

[Half Fast]

I know this is a bit of a derail, but I still find it interesting, so forgive me.

“It appears your calculation of KE_53/KE_48 = (53^2)/(48^2) = 1.2207 is imprecise. Any idea why that may be so?”

The answer, in part, still appears to be wrong:


You're correct, my calculation for KE_53/KE_48 = (53^2)/(48^2) = 1.2207 is imprecise. The actual value for this calculation is 1.2653.

The reason for this discrepancy is that I made an error in my initial calculation. Instead of calculating (53^2)/(48^2), I accidentally calculated (53/48)^2. This gave me a value of 1.2207, which is incorrect.

To correct this, the correct formula for the ratio of kinetic energies should be:

KE_53/KE_48 = (53^2)/(48^2)

Using this formula, we get:

KE_53/KE_48 = (53^2)/(48^2) = 1.2653


Weird how in its attempt to give human-sounding answers it appears to make human-like mistakes!

Even weirder that (1) it’s error got worse and (2) it’s explanation for the error is also an error. A^2 / B^2 does equal (A/B)^2.

Don’t trust this thing for math. Or anything else that’s important. And that’s AI in a nutshell.