We piston-twin pilots are in far deeper kimchee than our brethren who fly jets and large turboprops certified under FAR Part 25. Not only do our turbine-propelled colleagues have vastly better engine-out performance at their disposal, but the FAA also mandates that they be provided with detailed guidance—including balanced field length requirements and specific decision speeds—to tell them precisely what to do if an engine fails at any point. Unfortunately, the FAA requires no such guidance be provided to pilots of lesser aircraft certified under FAR Part 23 or its predecessor, CAR 3.
Study the emergency procedures section of any piston-twin POH and you’ll find precious little hard data on what to do if you lose one on takeoff. Can you abort without running off the end of the runway? Can you feather and keep flying without hitting the trees? The book doesn’t really say much more than “You’re on your own…good luck…be sure to write and let us know how it all came out.”
Because of the single-engine performance of most piston twins, it has long seemed to me that we poor pilots need specific procedural guidance for handling engine-outs even more than those well-heeled bizjet drivers do.
How Much Runway?
Pilots of Part 25 airplanes are required to calculate how much runway length is needed to guarantee that, in the event of an engine failure at any point during the takeoff, there is either enough runway left to abort the takeoff without running off the end or enough to clear all obstacles while climbing out on the remaining good engine. If the actual runway length is shorter than this calculated “balanced field length,” the pilot may not take off without first resolving the deficiency—generally by reducing takeoff weight, waiting until the OAT cools off, or both.
The POH for a piston twin does not provide data for calculating balanced field length. The data you will find in the POH includes:
- Takeoff distance over a 50-foot obstacle
- Landing distance over a 50-foot obstacle
- Accelerate-stop distance
- Accelerate-go distance
Doesn’t this give us the data we need to plan a safe takeoff? Well, not exactly.
The “takeoff distance” chart is fine so long as nothing hiccups during the takeoff. But if something does, you will most assuredly get up close and personal with that proverbial 50-foot obstacle.
The “accelerate-stop distance” chart tells you how much runway you’d need to accelerate to published liftoff speed, lose an engine at that point, and then chop and stop without running off the end. That’s certainly valuable data. But what happens if you lose an engine shortly after liftoff—say when you’re 10 or 20 feet in the air? You might well be too low and slow to fly away safely on one engine, and too high and hot to stop on the remaining runway. See the problem?
The “accelerate-go distance” chart tells you how much runway you’d need to accelerate to published liftoff speed, lose an engine at that point, feather the prop on the dead engine, and then continue the takeoff and clear a 50-foot obstacle. Now surely that’s enough runway to guarantee a safe takeoff no matter what happens, right?
Sure, it is. Actually, it’s more than enough—and that’s the problem. Accelerate-go distance gets ridiculously high when the temperature is hot or the field elevation is high (or both). In fact, if you look at the accelerate-go chart in your POH, you’ll see that this distance is often infinite, especially for normally aspirated twins.
An infinite accelerate-go distance means that if an engine fails at liftoff speed, the airplane will never climb out of ground effect, no matter how long the runway. Even if the distance is not infinite, hopefully none of you would ever attempt to continue a takeoff if an engine fails at liftoff speed!
Poor Man’s BFL
So how can we be sure that if an engine fails at an inopportune time, we have enough runway remaining either to chop and stop without running off the end or feather and fly without hitting those 50-foot trees at the end? That data isn’t in the POH but it should be, and it’s easy to calculate. All you need to do is look up two numbers—takeoff distance over a 50-foot obstacle, and landing distance over a 50-foot obstacle—and then add the two together to arrive at a poor man’s balanced field length. You’ll find that this number almost always falls somewhere between the accelerate-stop and accelerate-go distances.
What’s the significance of this calculation? It’s roughly the distance required to take off, climb to an altitude of 50 feet above the runway, then chop the throttles and land on the remaining runway. If the engine fails below 50 feet AGL you should be able to chop and stop with room to spare, and if it fails above 50 feet AGL you’re in a halfway decent position to feather and fly.
It’s not exactly the manufacturer-demonstrated, FAA-blessed balanced field length that I’d like to see, but it’s as close as we’re likely to get with the data we do have.
Try to Fly, or Chop and Stop?
Now, all this BFL stuff is lovely for preflight planning purposes, but what a pilot really needs is specific guidance on exactly what to do in the heat of that awful moment of realization that an engine has failed. Is it better to try to fly, or to chop and stop?
The turbine boys have a specific decision speed—V1—that provides an unequivocal, unambiguous answer to this question. If an engine fails below V1, chop and stop. If it fails above V1, continue the takeoff (and then land as soon as practicable, of course).
Permit me to offer a simple framework for making such go/no-go decisions. You’ll observe that it is somewhat spring-loaded toward chop-and-stop over feather-and-fly, and is predicated on the following prime directive: “It’s always better to go through the fence at 50 knots than to hit the trees at 120 knots.”
In that spirit, I offer three decision rules for engine failure on takeoff in a piston twin:
- If an engine fails on takeoff before reaching Vyse (blue line), flying is not a viable option.
- If an engine fails on takeoff and the gear is still down, flying is not a viable option.
- If an engine fails on takeoff at less than 50 feet AGL or below the height of obstacles at the departure end of the runway, flying is not a viable option.
The prime directive says that we should always chop-and-stop unless we’re darn sure we can feather-and-fly safely. The three rules state that to be darn sure, we need to have reached blue-line airspeed with the gear up (or at least on the way up) and an altitude of at least 50 feet (or higher if we have big obstacles ahead). Unless all these conditions are met, the fence is our best bet.
Why’d It Quit?
But what if the engine quits at a much safer altitude? Would you still shut it down immediately? Most likely not. Hopefully, you’d realize that you had enough altitude (and therefore enough time) to do some troubleshooting first. After all, it would be a real shame to shut down a perfectly good engine that could have been brought back to life simply by switching tanks, turning on an aux fuel pump, or switching off a misfiring magneto. If there’s time, try to figure out why it quit.
First, check the oil pressure gauge. If it indicates a loss of oil pressure, especially if confirmed by high oil temperature or the presence of oil on the engine nacelle, shut down the engine and feather the prop promptly to preclude seizure or fire. Otherwise, proceed to troubleshoot by checking the other three things that an engine needs to produce power: fuel, air, and spark.
Most engine failures are due to fuel starvation. This can be caused by lots of things: empty fuel tank, plugged fuel vent, mis-positioned fuel selector, failed engine-driven fuel pump, iced-up manifold valve, etc. But no matter what the exact cause, fuel starvation will show up on the flow gauge as little or no fuel flow. Conversely, if the fuel flow gauge reads normally, you can be pretty sure that lack of fuel is not the culprit. That leaves lack of air or lack of spark.
Lack of air will show up as sharply reduced manifold pressure, and there’s really only one thing the pilot can do about it: use alternate air. If alternate air does not restore the manifold pressure, a turbocharged engine should probably be shut down because of the possibility that the loss of manifold pressure is due to an exhaust system failure.
If fuel flow and manifold pressure are both okay, it’s probably an ignition problem. If both mags are on, try turning them off one at a time. If the engine comes back to life on just one mag, chances are the other one has gone berserk, so leave it off.
Just remember OFAS: oil, fuel, air, spark. If there’s time enough to troubleshoot, this is a logical sequence.
Time to Troubleshoot?
Whether or not there’s time to troubleshoot a failed engine is usually a function of altitude. We’ve already said that if an engine fails at 150 feet AGL, it’s no time to start playing diagnostician; just feather it focus on Job One—flying the airplane. On the other hand, if you’re cruising along at 5,000 feet when an engine quits, there’s typically plenty of time to figure out that you ran a tank dry, or your induction air filter is iced up, or your Golden Retriever got his paw snagged in the fuel selector. You’ve got airspeed and altitude to spare, so why not take the time to try to get the engine running again?
But what about engine failures that occur between 150 feet and 1,500 feet? It’s helpful to have some pre-defined rules to help decide whether to troubleshoot or not but, as usual, the POH doesn’t help.
Here’s a decision framework I like: If an engine quits below 50 feet AGL, chop and stop. If it quits between 50 and 500 feet AGL, don’t try to diagnose—just secure the inoperative engine and land as soon as possible. If it quits between 500 and 1,500 feet, do a “quick flow” troubleshooting pass from memory—oil, fuel, air, spark (OFAS)—but if the problem isn’t obvious within 15 seconds or so, give up and shut ‘er down. If you’re above 1,500 feet, you’ve got time to troubleshoot, so pull out the checklist and have at it.
If you don’t like my choices of altitude, pick your own. The main thing is to have a plan.
MIKE BUSCH is a CFIME and A&P/IA. EMAIL [email protected]
