Engine Fail/ Partial Engine Fail On Takeoff

[Airracer01]

Yeah, it's weird that we tend to almost exclusively practice complete engine failures, rather than partial engine failures. In my opinion, the latter take more ADM. The former is more a matter of airmanship and quick judgement. My own instance of a partial engine failure was at about 600ft AGL during sunny-day cruising, which led to what proved to be sufficient power to keep me level, but with rapidly rising oil temperatures which hit the limit of the gauge. The surrounding fields were conductive to landing and so the decision became one of putting it down immediately vs making a paved strip with maintenance personnel perhaps 5-10 minutes flight away. In my mind, I was looking at engine failure at any moment but I was also thinking about fire (not ideal in a fabric aircraft with wood spars). Ultimately, I elected to continue flight to the paved runway, with continual readiness to immediately pitch down for a controlled forced landing the moment the engine stumbled and to conduct a rapid emergency landing at the first indication of fire. I was able to put it down safety, with some scorched oil and bubbled paint on the engine cowling. It was certainly a focusing experience.

I would say that is a hell of a lot of ADM. I would think erring to the side of caution and not risking it anymore and putting it in a field would be a good option, but obviously your option was better that day! As you said being ready for a forced landing if she quit, or an emergency descent if it was on fire. Thank you for the response.

 

[Airracer01]

Brief every takeoff. "If I lose power prior to point A, chop (power) and drop (back down to the runway). Above/after that point, we'll land at *location* up until we reach *altitude* at which point we can turn around to the (right/left, whichever is into the wind) and land on the runway again."

The Aerial view in ForeFlight or Satellite in Google Maps is a good way to look for clear areas to set down shortly after takeoff in unfamiliar locations.

Also, I'm not a big fan of using 1000 feet blindly for the turn-back altitude. It really depends on the situation: A strong headwind can completely change the equation, for example. So, learn your plane thoroughly, and come up with a workable plan for the situation you're in.

And after you brief the takeoff, fly what you briefed.

I’ve had the scenario with a strong headwind and very low cold DA, turning crosswind by the end of the runway. If I have a complete fail there, I don’t know how one would maneuver for an impossible turn at that point, considering a short runway.

 

[Airracer01]

No.

First, while flying the airplane and looking for where you will be landing, unlock your door. This may well save your life if you are fortunate enough and skilled enough to land safely.

Second, drop the flaps (and the gear, if you need to), while you fly the airplane and proceed with “first”,above. You may not have time to reference your airspeed more than once or twice. This will give you the “extra” airspeed you need to avoid stalling and falling, and a cushion when you land on uneven terrain.

Third, fly the airplane until you are no longer moving. Then, exit the airplane. On an unprepared surface, fuselage twisting and/or airframe damage may make it impossible to open a locked door.

If Murphy truly rules your day, you will not have time to consult or to use a checklist. You will likely be reacting to a string of unfortunate events completely and utterly unfamiliar to you. If you do not react immediately and correctly, you will not survive the event.

Practicing loss of engine will only go so far. Thinking through multiple engine loss scenarios is just as important. Hangar fly. Chair fly. Daily, unless and until you have the experience, recency and currency in your aircraft to be able to do it without having to think about what to do next.

Good luck. Fly smart.

You definitely won’t have time to look at a checklist. If you don’t fly the airplane, you’re not going to make it like you said. Definitely no time for paper checklist, my question more or less is would there be time for a flow.

Chair fly, thinking through different scenarios daily is a go to for sure. Thank you.

 

[MauleSkinner]

To me, it’s a lot like an engine failure in a light twin…if you have enough performance to head for an airport, it’s probably worth doing. But it’s a constant evaluation process, and as soon as you realize you don’t have the performance anymore, you commit to landing off airport.

But that’s also not something we go out and train in multiengine airplanes.

 

[Airracer01]

I’m seeing a lot of blind 1000 feet turnback briefings too. I guess the assumption is that we have lots of choices at pattern altitude. When giving instruction, when I hear this, I’ve begun to ask, “how do you know 1000’ will work?”

Recently, I decided to play with this a bit…

ultimately, this is the best we can do. And to remember that the “best” place to put down may not be av”good” place to put down,

@Airracer01 Thank you for bringing this topic up! A little more than a year ago I had a similar situation happen to me where I lost a magneto on takeoff. I detailed the entire experience in a Lessons Learned post here. It was a learning experience, for sure and I don't recall ever having had a discussion with any CFI about partial failures - only total failures. To that end, most failures are due to fuel exhaustion just like a new Commerical pilot did here a few months ago...so don't run it out of gas and you'll most likely stay in the boundaries of partial failures. Two out of our three club planes here have had partial failures in the 2 years I've been a member. The Skyhawk had an intermittently sticking valve and most recently, the Skylane decided to give up a valve and a cylinder in the same takeoff roll. In all of these cases, the plane still flew, still gained altitude (albeit at a slower rate than normal) and all made it back home safe. I'd say the biggest lesson here which keeps getting restated is to fly the plane first.

Like you, if I'm going to be departing an unfamiliar airport I'm doing a quick scan on Google Earth to check for any optimal crash landing sites. I use that term intentionally because there are almost zero suitable places to land off-field in New England unless you're along the beach. Just take a look around the three fields I fly out of here...CQX - If you're out of 6 you have a golf course that's pretty open, if you're out of 24 you have Hardings Beach. HYA - Hopefully you can make half of an impossible turn and get the crossing runway no matter which runway you're out of because it's pretty much your only option. 5B6 - We'll help you pick up the pieces of your plane out of the trees or the cranberry bogs if you're lucky enough to make those. I never thought I'd miss the flat, farm field landscape of middle America but when the engine burps, I'll sure wish I had miles and miles of Texas.

No I know that feeling well. My home airport is surrounded by city and it’s go in the creek, a soccer field if there are kids in it, or dodge power lines and go on a road. I’ll have to give your post a read! Thanks for the response!

 

[Airracer01]

To me, it’s a lot like an engine failure in a light twin…if you have enough performance to head for an airport, it’s probably worth doing. But it’s a constant evaluation process, and as soon as you realize you don’t have the performance anymore, you commit to landing off airport.

But that’s also not something we go out and train in multiengine airplanes.

I remember watching a video on YouTube where a lady in her Piper Cub had a partial failure. She had airspeed under control and was going towards her nearest airport. Was maintaining altitude until it started to lose more power and then committed to landing off.

 

[flyingcheesehead]

I’m seeing a lot of blind 1000 feet turnback briefings too. I guess the assumption is that we have lots of choices at pattern altitude. When giving instruction, when I hear this, I’ve begun to ask, “how do you know 1000’ will work?”

Generally, we don't. It's just an easy rule of thumb. There are a LOT of variables:

Aircraft climb performance vs. glide performance - And even on the same airplane, this will be affected by variables like DA.
Headwind
Crosswind

I’ve had the scenario with a strong headwind and very low cold DA, turning crosswind by the end of the runway. If I have a complete fail there, I don’t know how one would maneuver for an impossible turn at that point, considering a short runway.

Exactly. The "Wow" moment in Mark's video above shows that. If you have a strong headwind and you can crank the plane around fast enough, you may end up overshooting if you try to land back on the departure runway. It's not out of the realm of possibility to have a strong enough headwind that it's best to make a full pattern by the time you get to 1000' AGL.

As far as the variables above, a plane that climbs really well (speaking of angle here, not just rate) should be able to turn back sooner than one that climbs at a lower angle. Same for a plane that glides really well. The more headwind you have, the sooner you can turn back since you won't have gone as far from the runway and you'll have a tailwind when you get turned around. The more crosswind you have, the less of a turn you need to make - Assuming you turn into the wind, you may be able to make only a 180 and have the wind blow you back to the runway centerline.

One other thing to note: If you're flying a glass panel - Anything that has the Flight Path Marker - You have a GREAT reference for where to point the plane. Generally speaking, with a decent crosswind you will NOT want to point your nose at the runway, as you'll fly a curved path to it due to the crosswind. If you put the FPM right on the runway threshold, you'll fly straight there, which is a slightly shorter path. In addition, if you put the FPM on the threshold and you're maintaining speed, you'll make it to the runway, whereas if you put the FPM on the threshold and you're slowing down, you're going to come up short. Knowing that so far in advance is incredibly useful and lets you come up with a new plan while you still have other options. "Bad news early is good news."

Recently, I decided to play with this a bit…

Cool! I'm a fan of doing some of your own test piloting.

I'm curious how much an idling engine affects things on various aircraft. I know that when we were doing the initial testing of the glide ring in ForeFlight, I pulled my engine to idle when I was at about 10,000 feet and the glide ring hit the airport. I ended up putting my gear out at 6,000 feet and flying a complete traffic pattern because of how high I ended up. I'm sure that my plane isn't THAT much better than the book when the fan is really off!

I really would like to do some actual engine out testing somewhere with a huge runway surrounded by lots of fields and no people. Take it up to 14,000 repeatedly. Test out things like whether it's better to hold altitude until Vg or pull up to bleed the speed off quicker and get a little extra altitude out of it, effect of prop and throttle levers on actual glide performance, effect of windmilling prop, glide ratio in various configurations to determine when it's safe to add flaps and gear, etc... But I'm not going to bother doing this until I've got the right avionics to log EVERYTHING (including engine data, which my plane currently logs separately).

In addition, some testing related to the impossible turn would be informative with some math to start out. First, calculate the theoretical amounts of time to turn 180 degrees and maybe every 10 degrees after that up to 220, the stall speeds and turn radii at different angles of bank, and how much time it really takes to roll in and out of those turns. Then, go up and see whether performance matches expectations, and finally figure out how the variables above affect it all. Maybe even put together a spreadsheet where you could input wind speed and direction, runway heading, and maybe runway performance as calculated in ForeFlight. Then, calculate when it's best to stop on the runway, when you'll need to keep going ahead, what altitude is the minimum for the impossible turn, and the optimum bank angle and rollout heading to achieve the best performance on the turnback.

Yes, I'm a nerd.

ultimately, this is the best we can do. And to remember that the “best” place to put down may not be av”good” place to put down,

Sticking to your guns when the best option is non-ideal is difficult. Gotta be Sully and put it in the river anyway.

 

[MauleSkinner]

One other thing to note: If you're flying a glass panel - Anything that has the Flight Path Marker - You have a GREAT reference for where to point the plane. Generally speaking, with a decent crosswind you will NOT want to point your nose at the runway, as you'll fly a curved path to it due to the crosswind. If you put the FPM right on the runway threshold, you'll fly straight there, which is a slightly shorter path. In addition, if you put the FPM on the threshold and you're maintaining speed, you'll make it to the runway, whereas if you put the FPM on the threshold and you're slowing down, you're going to come up short. Knowing that so far in advance is incredibly useful and lets you come up with a new plan while you still have other options. "Bad news early is good news."

The analog version of that is a bug on the windshield.

In addition, some testing related to the impossible turn would be informative with some math to start out. First, calculate the theoretical amounts of time to turn 180 degrees and maybe every 10 degrees after that up to 220, the stall speeds and turn radii at different angles of bank, and how much time it really takes to roll in and out of those turns.

Don’t forget about altitude loss in the turn. typically, a steeper bank will lose less altitude in the turn than a shallower one even though it has a higher descent rate, because you spend significantly less time in the turn.

Unfortunately what you end up with is “make the steepest turn you can make without stalling.”

 

[flyingcheesehead]

The analog version of that is a bug on the windshield.

Yes. If you can perfectly keep the runway in the same spot... But it's a whole heckuva lot easier with the FPM, because you should be able to target the rollout of the turn almost perfectly.

Don’t forget about altitude loss in the turn. typically, a steeper bank will lose less altitude in the turn than a shallower one even though it has a higher descent rate, because you spend significantly less time in the turn.

Unfortunately what you end up with is “make the steepest turn you can make without stalling.”

Yup... That's what gives you the smallest turn radius, too. However, I can see a scenario where, if you have enough headwind and enough crosswind, that you don't need every bit of altitude to make it back to the runway, and where even an exact 180º turn done quick enough will have enough crosswind where you'd cross the centerline prior to being back on the ground. If both of those are true, a shallower bank angle might end up working better.

 

[MauleSkinner]

Yup... That's what gives you the smallest turn radius, too. However, I can see a scenario where, if you have enough headwind and enough crosswind, that you don't need every bit of altitude to make it back to the runway, and where even an exact 180º turn done quick enough will have enough crosswind where you'd cross the centerline prior to being back on the ground. If both of those are true, a shallower bank angle might end up working better.

I was assuming the “minimum altitude,” “optimum bank angle”, and “best performance” that you stated in your post.

 

[flyingcheesehead]

I was assuming the “minimum altitude,” “optimum bank angle”, and “best performance” that you stated in your post.

Yes - And while "as steep as you can without stalling" is probably the way to go at least 95% of the time, it's not necessarily 100%.

 

[orca64]

You definitely won’t have time to look at a checklist. If you don’t fly the airplane, you’re not going to make it like you said. Definitely no time for paper checklist, my question more or less is would there be time for a flow.

Chair fly, thinking through different scenarios daily is a go to for sure. Thank you.

The answer is it depends on where (altitude agl) you are when you lose the engine.

At 500-600 feet, you may have less than 60 seconds from the start of your adventure to its ultimate end. If you’re very lucky you’ll find a good place to land; if not, you’ll need to make a place to land.

No time for any checklist.

Probably not very realistic to try to pre-program or practice any “flow”, except for those Automatic items we all do when/if our engine starts running rough—or not at all. What kind of a flow, would depend on what kind of a failure and the geography that you’re flying over. Way too many permutations there. Regardless, concurrently, fly the aircraft.

At very low altitudes you will find yourself reacting to situation and circumstance. You will live or die based upon your ability to correctly analyze your situation and take proper action.

Good luck. Surely, we’re all counting on you.

 

[MauleSkinner]

Yes - And while "as steep as you can without stalling" is probably the way to go at least 95% of the time, it's not necessarily 100%.

And since “as steep as you can without stalling” is probably the way to go 95% of the time, I would definitely not recommend anything beyond “ahead of the wings, into the wind” until you’re sure you’re well into the remaining 5%.

 

[KE5BM]

Good luck. Surely, we’re all counting on you.

I see what you did there

 

[orca64]

I see what you did there

That’s not important now.

 

[midlifeflyer]

I really would like to do some actual engine out testing somewhere with a huge runway surrounded by lots of fields and no people

One of the comments to my video was from a simmer who found it interesting. In reply, I wondered whether the flight dynamics in something like X-Plane or even a Prepar3d-based Redbird were good enough to do a good engine out simulation.

 

[Ed Haywood]

Good luck. Surely, we’re all counting on you.

Don't call him Shirley.