Right from the beginning, I knew that we were cutting it tight. As I steered the Citation Mustang onto the runway, the temperature in Aspen was cool for a summer day, but at an elevation of 7,800 feet, we just barely had enough pavement to make a safe departure if we lost an engine at the wrong moment. I also knew that we couldn't make the required climb gradient on the departure procedure without both engines. When it's hot, and you're high and heavy, losing an engine can really get your attention.
The tower cleared us for departure, and I pushed the throttles to full takeoff power. No doubt about it, the acceleration felt anemic, and with a clear view of the end of the runway rushing my way, we finally reached rotation speed. It seemed like we just barely cleared the end of the runway as I raised the gear and retracted the flaps. It almost seemed too easy as we climbed into the overcast at 9,500 feet, and that's when it happened. Power to the left engine simply rolled back to zero. The panel was full of flashing lights as I ran through the memory items and re-trimmed the rudder. I keyed the mic to announce that we had an emergency, and that we were going to attempt a return to the airport. To my surprise, the tower announced that they had a disabled aircraft on the runway, and that I'd be unable to return. Heck, I just departed—how did that happen? There was no time to worry about it now. There were mountains all around, and our climb rate had fallen to only a few hundred feet per minute. I needed a plan to stay out of the rocks.
Staying on the departure procedure was out of the question, so I pulled up the terrain page on the G1000. I could clearly see the pass that leads northwest around to Eagle. So, I turned to stay out over the valley to avoid the peaks that towered above. With that taken care of, I worked through the engine shutdown checklist while dividing my attention to make sure that I stayed over lower terrain. It didn't take long before I had gained enough altitude to see that I could easily make Eagle. At that point, the lights came on and my instructor said, "Good job, you'd be surprised at how many guys stay on the departure and wind up crashing into a hillside." The level-D sim is so realistic that it's a bit disorienting to jump back to reality at the end of a training session like this one. Still, it's one of the best ways to experience and prepare for the realities of an engine failure on departure.
Jet Departure Considerations
So, what are the things that a Part 91 operator has to consider in order to make a safe departure in a multi-engine light jet? It basically comes down to the same three issues that we had to consider when we first learned to fly. Is there enough runway, what do you do if the engine fails at some point during the takeoff roll, and what are you going to do if the engine fails at the worst possible moment? When you transition into the world of multi-engine jets, you'll find that these considerations get a bit more complicated than when you were flying a piston single. Turbofans are tremendously reliable, but don't forget that when you have two of them, you always have twice the statistical likelihood of a failure compared to an aircraft with only one of the same engines. It's a good thing to remember that even with all that reliability, jet engines do occasionally fail, so don't wave it off as something that won't happen to you.
We all learned as students that FAR 91.103 requires that every pilot needs to be familiar with the runway lengths at the airports of intended use along with the takeoff distance data contained in the aircraft POH. In a multi-engine jet, the takeoff distance is determined by computing two numbers. First, we compute the distance from brake release to where the aircraft reaches an altitude of 35 feet with both engines running at full takeoff power and multiply by 115%. Second, we figure out the distance needed from brake release to where the aircraft reaches an altitude of 35 feet after an engine failure occurs at the V1 decision speed. The decision speed (V1) is the speed where you determine whether or not to take off after an engine failure. Lose the engine below V1, and you do everything possible to stop on the runway. Lose the engine at or above V1, and you accelerate to rotation speed (Vr) and take off. V1 is computed for every takeoff using takeoff weight, aircraft configuration, elevation and temperature. The only caveat is that these distances are computed for a dry runway. If the runway is wet, the 35-foot requirement may be reduced to 15 feet. In either case, the takeoff distance is then given by the larger of the two numbers calculated by each method. Keep in mind that these calculations need to take into account the effects of wind and runway slope, as well.
The takeoff distance doesn't necessarily tell you how much runway is needed. To determine that, you also need to consider the accelerate-stop distance. That's the distance from where you apply maximum takeoff power, release the brakes, accelerate and then lose an engine at a speed just below V1, to where you come to a stop after you close the throttles, deploy the speed breaks and slam on the brakes. Just be aware that a wet or icy runway will have a really big impact on the accelerate-stop distance. That's one reason that a really long runway is needed when the surface is contaminated.
So, the required runway length is determined by the largest value of either the takeoff distance or the accelerate-stop distance. There are a few other factors that can come into play to complicate matters, but this is how most light jet manufacturers determine runway length requirements. If all this sounds like a bit of chore to compute from the books for every takeoff, it is. Fortunately, computers, smartphones and tablets are really good at this kind of calculation, so most manufacturers make performance calculators available to simplify the process. In addition, most aircraft checklists include tables using simplified criteria to make it easy to look up V-speeds and runway lengths for a range of standard conditions. So, in reality, this process is quite easy and only requires a few minutes to accomplish. What you get in return is the assurance that even if you lose an engine at the worst possible moment, you'll have enough runway to either stop or get airborne.
The Initial Climb
Assuming that we have enough runway to safely take off, the next thing we need to know is whether we have enough performance to safely climb out. A jet takeoff is divided into four segments, so we'll need to look at each segment to see if we can meet the minimum SE climb requirements set by the FAA. The first climb segment starts at 35 feet and ends when the gear is up and the aircraft achieves V2. V2 is defined as the speed that provides the best climb gradient with the most critical engine inoperative. It's essentially Vxse, and it's the speed that needs to be maintained with a failed engine. With both engines running, even very light jets will blow right through V2 so fast that the first climb segment normally doesn't last very long.
It's the second climb segment that causes a lot of the trouble with an inoperative engine in a jet. The second segment starts as soon as the gear is retracted and extends up to 400 feet. The FAA requires jet aircraft to be able to maintain at least a 2.4% climb gradient up to 400 feet with one engine inoperative. That's only 24 feet per 1,000 feet, so it's not very steep. At a V2 speed of, say, 100 knots, that's only 242 feet per minute, but on a hot day in a heavy loaded jet departing from a mountain airport with only one engine turning, that may prove to be a very difficult proposition. Even at low elevation, there are a number of airports with very small hills off the end of the runway that will start to look pretty big if an engine fails right at the worst moment. Remember that the normal minimum climb rate required to meet TERPs requirements on an instrument departure is 200 feet per nm. That works out to be a 3.3% gradient, so at the very minimum, you won't meet TERPs terrain clearance. If there's anything sticking up above the departure end of the runway within about two miles, you're going to want to be able to climb at a higher than minimum allowed gradient—no matter what. Normally, that means departing at a lighter weight or with cooler temperatures. If you have plenty of runway, you can often achieve a better climb gradient with zero flaps and/or by adding a bit of speed before rotation to more quickly accelerate to V2. Another thing to understand is that the actual climb gradient should be determined from the point where the aircraft leaves the ground—not at the end of the runway. Some electronic performance calculators take this into account to allow more operational leeway in the face of obstacles off the end of the runway.
So, why is this stuff so important in a jet? In a single, the FAA is okay with the idea that the aircraft is going to come down after an engine failure right off the end of the runway, but they're not so happy about that idea when you move into a twin jet. There are a couple of good reasons for that. First, twin jets tend to weigh more and go faster than most singles, which means that there's a lot more kinetic energy to cause mayhem in the event of a crash. Second, with only a couple of exceptions, most certified singles are required to have a maximum Vso (stall speed in the landing configuration) of 61 knots. The idea is that a single could come down with considerably less destructive energy—both for the occupants and for whatever the plane might hit. Twin jets have no such stall-speed restriction, so the FAA requires that if a jet loses an engine, it should be able to safely continue. There are certainly cases where this restricts the operations of a twin jet with respect to, say, a single engine turboprop—even though the jet will probably outperform the turboprop with both engines running. However, this limitation contributes to a safer operation relative to what any single can achieve in an engine failure.
Even if we can get ourselves up to 400 feet after an engine failure, we still need to continue the climb. The flight path from 400 feet to 1,500 feet AGL combines the third and fourth final climb segments. In the third segment, we accelerate to the final segment speed, which is the speed needed to achieve the best possible single engine angle of climb, normally with the flaps up. The fourth climb segment is allowed at a minimum climb gradient of 1.2% to 1,500 feet AGL. Again, this isn't very steep! It's only 12 feet up for every 1,000 feet forward, so you might cover another 10 miles to get to 1,500 feet above the runway. If that's all the performance you can eke out of your jet with a failed engine, and you're not in Kansas, things could get pretty exciting. So, checking aircraft performance and the minimum required performance on a departure becomes more then academic in the event of an engine failure.
Go Or No Go?
If you're making an instrument departure, just how restrictive are these requirements? The AIM makes it clear that ODPs and SIDs assume normal aircraft performance with all engines operating, so Part 91 operators are legal to depart as long as you can meet the climb requirements with everything working. Section 5-2-8.e of the AIM also states that during preflight: "Each pilot, prior to departing an airport on an IFR flight should:
(a) Consider the type of terrain and other obstacles on or in the vicinity of the departure airport;
(b) Determine if an ODP is available;
(c) Determine if obstacle avoidance can be maintained visually or if the ODP should be flown; and
(d) Consider the effect of degraded climb performance and the actions to take in the event of an engine loss during the departure. Pilots should notify ATC as soon as possible of reduced climb capability in that circumstance."
It's your responsibility as pilot in command to figure out beforehand what you're going do if you lose an engine on the departure. Obviously, the weather plays a big part in that decision. If it's low IMC, and you're in the mountains, losing an engine on an ODP may mean that you can't make the required climb gradient, which could put you in the rocks. Simply assuming that nothing will fail probably wouldn't qualify as a valid plan, and it's likely that the FAA would take a similar view if you made that decision, got yourself in trouble and survived to talk about it. Departing under these conditions without the ability to comply with an ODP on a single engine wouldn't be very smart and might not be legal. On the other hand, if weather conditions permit easy terrain clearance (e.g. VFR), an engine failure would constitute an emergency, and it's fair to assume that you'd simply depart from the procedure (SID or ODP) to return to the airport. One point that's important to emphasize is that if you're on an ODP, the published altitude restrictions are necessary for obstacle clearance and/or design constraints. Compliance with these restrictions is mandatory and can't be lowered or canceled by ATC. The ODP isn't simply a "legal" requirement. It's there to help keep you from tangling with terrain no matter how many engines are running.
Most of the time, the go/no-go decision in a light jet is simple, but when weather combines with hot, high or heavy considerations, it's important to pay attention to single-engine performance limitations. You can find more detailed information on this subject in Advisory Circular AC 120−91, Airport Obstacle Analysis, the "Departure Procedures" section of chapter 2 in the Instrument Procedures Handbook FAA−H−8261−1 and in Everything Explained for the Professional Pilot by Richie Lengel. Safe travels!
John Hayes is type rated in the Citation 500 series and Mustang and is an ATP, CFI, CFII, & MEI with over 4,600 hours in numerous airplanes. A founder and past president of TBMOPA and the Citation Jet Pilots, he enjoys flying aerobatics in an Extra 300L.