[Editor's Note: Recently two flight instructors wrote a book on how to use Microsoft Flight Simulator X (FSX) to enhance pilot training and to provide sim-only pilots a guide to making their flying more realistic. AVweb is reprinting several chapters from this book, the first of which was Chapter 13 -- Weather. To download the FSX files they refer to here, visit the publisher's Web site and click on Downloads.]

There's a clich in multiengine flying that says the purpose of the remaining engine is to carry the airplane to the crash site. That's only half in jest. Losing an engine in the cruise section of flight is usually not that big a deal -- well, OK, it's always a big deal, but it's usually not dangerous. Losing an engine while climbing away from the ground is often deadly. (See "Accident Report: Know Your Priorities" at right.)The Beechcraft Baron is one of the few light twins that will climb reasonably well on one engine. You've got this in your favor. The real Baron, especially the older, straight-tail Baron, has a small vertical fin and rudder for its power. As you'll see in this chapter, a small vertical surface makes it more susceptible to loss of control when an engine unexpectedly quits.

When an engine fails in a twin, everything gets worse. There is a 50 percent reduction in power that, as we discussed in Chapter 21, will result in a 70-90 percent reduction of climb rate. What we have not looked at as much is that this power is also asymmetrically distributed. This causes a host of problems. (Unless the twin is designed differently -- see "Centerline Thrust" at right.)

Let's assume the engine on the left wing quits. The propeller that was being spun by the motor will keep spinning, but now it's the airflow passing over the propeller blades that is spinning the propeller. This creates significant drag on the left side of the airplane (see Figure 23-1). The drag will exacerbate the problem of lost lift. Not only do you have less lift to use for climbing, but you also have more drag to overcome, which means you need even more lift than normal to keep climbing.The drag is also off-center in that it's acting only on the left wing. This means the airplane will try to yaw to the left; that is, the nose will swing toward the dead engine. The thrust coming from the good engine makes this worse. Rather than pulling along the centerline of the airplane, the good engine is pulling from out on one wing. It will also try to yaw the airplane toward the dead engine.What's your reaction? You stomp some right rudder, of course. This is a good plan and will stop the nose from swinging, but that heavily deflected rudder is even more drag. Now you're sinking faster ... see how this is starting to add up?

If all that yawing wasn't bad enough, the airplane will also try to roll toward the dead engine. Part of the roll is just a factor of the yaw. You already saw, when you first started your training, that just deflecting the rudder will start the airplane skidding in a yaw but will also start a roll in the direction of the turn.

The failed-engine scenario is a bit worse, though, because the wing area directly behind a running engine actually produces extra lift from the accelerated air flowing through the propeller. Kill the engine, and this bonus lift becomes inhibited lift because the windmilling propeller actually disrupts the airflow over the wing. Now you have significantly different amounts of lift on the two wings, which causes a rolling moment toward the dead engine (see Figure 23-2).You're ready for this too, so you turn the yoke to counter the roll. That's fine except the increased lift from the aileron creates some adverse yaw, which adds to the yaw toward the dead engine.Now you're sinking and yawing and rolling. You need more power to at least stem the altitude loss. The only way to do this is to bring the good engine up to full power if it wasn't there already. Of course, adding power to the running engine just makes all the discrepancies we just discussed even worse.

On the majority of propeller-driven, twin-engine airplanes, losing the left engine is more dangerous than losing the right engine. This is for the same reason that single-engine airplanes turn left when they're climbing: because the engine rotates clockwise from the pilot's point of view. You'll remember that this is called "p-factor," where the descending blade, which is on the right side of the airplane, has a greater angle of attack, and therefore a greater thrust, than the ascending blade on the left side of the airplane.On a twin, both engines have some asymmetrical thrust in a climb -- and, yes, the twin will wander left in the climb with both engines running -- but it's worse for the right engine.Picture the Baron from above (see Figure 23-3 below). The descending blade of the right propeller is farther from the center of the airplane and has more of a lever arm to apply its yawing force. So if the left engine fails and leaves you with only the right engine for the climb, you'll need more rudder force to keep flying straight ahead than if the right engine had failed and you were climbing using only the left engine. Because the failure of the left engine creates a more hazardous situation, the left engine is thought of as the critical engine.

Some aircraft have counter-rotating propellers. With very few exceptions, counter-rotating propellers both rotate toward the fuselage, so the yawing effect of the engine loss is at least minimized. On these airplanes each engine is equally "critical." They also have no turning tendency in a climb with both engines running.

As we said in Chapter 21, V_mc is defined as the minimum controllable airspeed with the critical engine inoperative and is important enough to be shown as a big red line on the airspeed indicator. Below V_mc there isn't enough rudder authority to prevent the yaw -- and, much more important, roll -- toward the dead engine. An airplane flying below V_mc with full power on only one engine might roll toward its dead engine and crash without being able to prevent it (see Figure 23-4).That definition glosses over some details, though. Many factors affect V_mc. The farther aft the center of gravity, the higher the V_mc becomes because the rudder has less of a leverage arm to counter the yawing motion. Having landing gear extended would lower V_mc because the gear would act like a vertical fin helping to stabilize the airplane. You can read "The Full Scoop on V_mc" at right for a link to the full list of what goes into V_mc, but for our purposes, we'll say that the published redline is the highest V_mc the Baron will see. Actual V_mc might occur at a lower airspeed, but by never being airborne below V_mc (in a real Baron anyway), we won't risk the V_mc roll.So, you might wonder why you can't just reduce the power on the good engine if the aircraft starts to roll. Wouldn't that stop the roll and let you level the wings? Sure it would. It would also stop you from climbing, which could be a serious problem with an engine failure near the ground. It would still be preferable to crash land with the wings level and some control than to hit inverted with one engine screaming at full throttle.There is a dangerous misconception, though, that develops in flight training with V_mc on this very question. Most of the light trainers that people use for flight training have low-power engines and big rudders. The V_mc roll is slow, and a power reduction combined with a healthy dose of rudder will easily stop the excursion. These trainers also have low stall speeds, so V_mc is well above the stalling speed of the airplane.Higher-performance airplanes, including the Baron, have higher stall speeds and a snappier V_mc roll. They are also more likely to be under their gross weight just because they can carry bigger payloads but often fly without all those seats filled. They might be close to the stall speed when the V_mc yaw and roll hits, and the yawing motion could stall the wing on the inside of the turn. This is the setup for a spin. Spins in many light twins are difficult or impossible to recover from.The moral to the story is to respect that redline on all real-world airplanes!

Exact engine-out procedures vary from aircraft to aircraft, but the right thing to do is generally the same in most light, twin-engine airplanes. This is the procedure you'll use in the Baron.

When an engine fails in flight, the nose will drop and the aircraft will yaw and roll. You'll instinctively do the first correct step: maintain control with pitch, rudder, and roll.More specifically, you'll pitch as necessary to maintain altitude (or your current climb or descent as appropriate), but you won't fly slower than a blueline of V_yse. If the airplane reaches V_yse, you'll pitch to maintain that speed and accept whatever rate of descent you get. You'll also use your rudder to keep the nose pointing straight ahead and whatever aileron input you need to keep the wings level for the moment.You just lost half your power, and you might not even be sure which engine it was. You'd be surprised how intuitive correcting your heading with rudder and roll can be. You don't think about it; you just do it. Since you're not certain which engine it is, you'll push both throttles to full power, followed by both prop controls to full power. This should give maximum power on the good engine and have no effect on the windmilling one.Next you'll move both mixture controls to max power. This might be full forward, but it might mean leaving them put. It's often not a bad idea to move them a bit forward no matter what just in case you had them too lean and that caused the engine stoppage.If you have gear and flaps extended, raise them now to reduce your total drag.

After checking to see that you're still flying straight and aren't about to hit anything, you can better your situation by banking the airplane slightly toward the good engine. Some folks will tell you to bank exactly five degrees. This is hogwash. You want to bank enough to eliminate the sideslip through the air caused by the deflected rudder. Although it's hard to tell where this point is without a yaw-sensing device on the airplane, you get a feel for it in the real plane. It's often just a couple degrees of bank, as shown in Figure 23-5.If you bank the right way, you'll feel the rudder pressure decrease, and the airplane will track straighter through the air. The ball in the inclinometer under the turn coordinator will be about 1/2 a ball width off center, but that's just fine.If you bank the wrong way, you'll feel that more rudder is needed. In fact, as you bank toward the dead engine, you're increasing V_mc at the rate of about 3 knots per degree of bank.

Now it's time to identify which engine is down, verify that you have it right, and then attempt to either restart the engine or secure it.Retard the throttle on the engine you think has failed. If you retard the wrong throttle, it will be immediately clear because you'll rapidly have no engine power at all. If the throttle has no effect, you can assume you have the dead engine identified.

Now you must take a look around and decide whether you have time to troubleshoot the problem. If you do, you'd attempt a restart just as you did in previous chapters. Let's assume you don't have that luxury. You're near an airport, barely holding altitude, and need to land. Pull the prop control for the dead engine halfway back, and verify you are moving the correct prop control both by looking at the throttle quadrant and by feeling that the available power didn't diminish. Now pull the prop control the rest of the way back to the feather position (see Figure 23-6). This will stop the propeller and turn it so it presents minimum drag. Your V_mc will decrease when this happens, and the airplane should climb better and need even less rudder and bank.Now verify which mixture control is for the dead engine, pull it to idle cutoff, and turn the fuel for the dead engine to off. (See "Crossfeeding the Fuel" at right for another consideration.)Take a moment now to look after your good engine. Open the cowl flaps as needed. Adjust the power, prop and mixture as needed. Prepare for a single-engine approach and landing.

You'll start your flights with the most benign of engine failures: failure during cruise flight. You're in a lightly loaded Baron and at 8500 feet en route from Cheyenne, Wyo., (KCYS), back to your base at Jefferson County (Jeffco) airport (KBJC) in Colorado. About a minute into the flight, your left engine will fail. Your job is to land at Jeffco safely.

The first step is realizing you even have a failure (see Figure 23-7). Since you're cruising with the autopilot on, you'll see the heading wiggle, but the autopilot will try to compensate. Start looking to find the problem. The prop-sync pinwheel spinning is a tip that something is up with the engines. You'll see from the left-side engine gauges that you have a serious issue.

Sophisticated autopilots can handle an engine failure well. This autopilot doesn't qualify, so disconnect it. You'll immediately need some right rudder. Watch your speed now. It's above blueline, so you're fine for the moment, but it will rapidly drop. When it hits 101, pitch to keep 101, and accept the slow descent. Push the throttle and prop controls full forward to get maximum power. Set your mixtures for max power, which probably means leaving them alone if you already set them for altitude.

Bank toward your right (good) engine about 3-5 degrees. Use enough rudder to keep the black inclinometer ball about 1/2 off center (see Figure 23-8). In the real world, this will reduce drag and reduce V_mc. It doesn't seem to have quite the right effect in FSX, but it's close enough to practice the correct procedure.

0)]Since you've got a little altitude and time, you can troubleshoot this engine. This is an urgent situation but not an emergency. Before you go any further, though, keep your good engine happy by opening the cowl flaps and periodically checking on it. Also keep heading for KBJC. As you can see in Figure 23-9, you've got a situation and will need to land sometime soon.We failed this engine with the Aircraft > Failures option, so there's no way it will restart. But engines stop for want of at least one of three things: fuel, air, or spark. Check those systems now by doing the following:

Note that you do not need to cycle the ignition switch to Start. The propeller is spinning in the wind. If you get the right conditions in the engine for combustion, the engine will start on its own.

1)]OK, it won't start. You'll cut your losses and secure it, as shown in Figure 23-10. Step one is to make sure you have the correct engine by pulling the throttle to half throttle and then to idle. When that doesn't have any effect, do the same to the prop control. When you pull the prop control fully aft, you'll be in the feather mode, and the propeller on the wing will stop. Remember you can press E and then 1 on your keyboard to make your joystick control the right engine controls. You will need to move the actual feather control onscreen or press Ctrl+F1 to pull the prop to minimum rpm and then press Ctrl+F2 to pull it further to feather.With the prop stopped, you'll turn off all the fuel to that motor: mixture to idle cutoff, fuel selector to off, any boost pumps off, and magnetos off.

The last step is to fly the visual approach with only one engine. It's not hard to do, but the key factor is managing drag. Right now, the Baron has its gear and flaps up. Put those gear down, and you'll be descending about 500 fpm without changing anything. The flaps will net about 400 fpm. Flaps and gear could be as much as 1000 fpm down when coming into this high-altitude airport. You'll have to plan this approach carefully.Oh, and not to add any pressure, but your single engine won't supply enough power to go around if you botch the approach. You have one shot to get it right.For that reason, pick a long runway for the approach so you can aim partway down the runway and have some cushion to overshoot or undershoot. If you have the option, your best bet is usually a long, straight-in approach. That way you can get a steady descent rate established and take it to the runway with few changes. If you have to fly a traffic pattern, fly with the good engine on the same side as the runway. That way you'll make all your turns into the good engine and maintain maximum control.If the airport has a tower, let them know you're a single-engine approach, and tell them what you want to do. They'll work with you and let you fly right traffic or a long straight-in and get other airplanes out of your way.You should be over the airport or nearby by now. Since FSX ATC can't handle special requests, you'll fly this one yourself.You'll try both a straight-in and a pattern. So you can get back to this point quickly, choose Flights > Save, and save this flight right now as "temp." Now turn eastbound, and use your GPS to get about 5 miles from the airport and turned back inbound to Runway 29R. If you do this efficiently, you should still be higher than 7000 feet MSL.Now get lined up with the runway, and lower your landing gear. See what this does to your rate of descent while maintaining 101 knots. If you're still looking good for getting to the runway, reduce the power by a few inches and see how well that works. Your aiming point shouldn't be the runway threshold but rather the thick, white 1000-foot marks down the runway. If you're still looking good, lower approach flaps, and see what that does.

2)]The goal here is a configuration with the gear down and whatever degree of flaps you can manage that gives you about 500 fpm down while still going 101 knots (see Figure 23-11). Ideally, you'll be at a few inches less than full throttle, too. That way you have some power to add if it looks like you're going to come up short.Each condition will be different. A lightly loaded Baron at sea level might be able to make a final approach with gear down and full flaps with less than full power on one engine. A heavy Baron at altitude may need to make the approach with gear and flaps up and full power on one engine until short final. The gear would come down only 20-30 seconds before touchdown. This is one of those places where flying becomes an art.

3)]Press Ctrl+; on your keyboard to reset the flight. You're back in the air with a feathered engine. Now you'll need to maneuver to enter a right downwind for Runway 29R. This is a bit trickier. You know when you put the gear down that you'll start coming down at 500 fpm. So where do you start? Maneuver to enter a wide downwind at pattern elevation of 6700 feet (see Figure 23-12). You may need to overfly the airport and turn to lose a bit of altitude before you enter the pattern.

4)]Fly the downwind in level flight with the gear up. When you get abeam your landing target of 1000 feet down the runway, lower the gear, and let the Baron descend, ideally about 500 fpm (see Figure 23-13). After you've come down 200 feet to 6500 feet, turn right for the base.

5)]As you roll out on base, make your judgment as to how high or low you are, and add flaps accordingly (see Figure 23-14). You're better off too high than too low. Remember as well that your aim point is not the threshold, but 1000 feet down the runway. Turn final when you're ready. Only lower flaps to approach or full if you're certain you have the runway made. At this altitude, it's unlikely you'll want full flaps, and landing with only partial flaps is fine.If you don't make it, well, be glad it's a simulator. Reset the flight, and try again.

Load the flight Chap_23_IMC_failure. You're at 10,000 feet flying from Cheyenne to Jeffco again, but the weather isn't so hot this time; there's rain and low ceilings all around. It's a good thing you have two engines ... oops, there goes one of them. We'll let you figure out which one. It might, or might not, be the same one we use in the following figures.

6)]Once you've figured out it won't restart, feather and secure it. Your first issue is that you won't be able to maintain 10,000 feet and stay above blueline (see Figure 23-15). In the real world, you'd tell ATC about your problem and get cleared to a lower altitude immediately. You can't do that in FSX, so just acknowledge when they tell you to climb, and ignore the requests to expedite.Soon you'll get a descent to 7200. You're going pretty slow, so if you want to keep the power up and just point the nose down, be our guest. Be sure you have full power on the good engine when you level off. You should be able to maintain altitude here at 7200. Once you're at 7200, choose Flights > Save, and save this flight as "temp." You'll be asked whether you want to replace the old temp. Click Yes.Follow ATC's vectors. When you intercept the localizer, adjust your throttle setting to get 110 knots. This is your normal approach speed, and in an abnormal situation like this, you want to keep as much normalized as possible. Load and activate the ILS Rwy 29R approach in your GPS for extra situational awareness too.

7)]When you see the glideslope needle fully center, start down the ILS, and then lower your gear. If this seems backward, it is, but we find getting that small boost in speed pointing down the ILS and then adding the drag of the gear helps avoid getting too slow on the approach (at least on the FSX Baron). You might not need to adjust the power at all, but if you do, adjust it just enough to maintain 110 knots. Keep the flaps up for this high-altitude approach until you have the runway made. In fact, you're probably best off not bothering with them at all (see Figure 23-16).

Flying an ILS like this is the preferable approach for the same reason as with the long, straight-in, visual approach. You have more time to gently adjust your parameters to get a stable descent to the runway. Alas, an ILS isn't always available, so you'll need to know how to fly a nonprecision approach sans engine as well.

8)]Press Ctrl+; or load the flight temp. Now you're back in the air en route to KBJC. Use the ATC window to request another approach, and get the GPS Rwy 29R for KBJC. The approach is essentially the same except for one scary tidbit. Once you drop the gear, you will not have enough power to maintain altitude up here. You'll fly the approach at 110 knots with the gear and flaps up (see Figure 23-17). This will allow you to level off at intermediate altitudes and the final MDA of 5900 feet. When you see the runway and have at least red over white on the VASI, you can lower the gear and slow to 101 for a final approach. If and when to extend approach flaps is up to you.All the engine failures you did in flight earlier were at a light weight. If you want some additional insight into how weight matters, take off in this fully loaded Baron and then fail an engine at altitude. You'll see how much more difficult it is to get back to pavement safely. (See "Mountain IMC" at right for another high-altitude challenge.) For the opposite effect, try any of the engine-failure flights at sea level. You'll find it much easier.

There isn't much to this one beyond the planning. The key is remembering that you want to maneuver to land with the good engine on the upwind side. This might mean crossing over an airport and coming back to land the other way. You already did a crosswind landing with much more power on the upwind engine compared to the downwind engine. Now all the power is on the upwind engine.Load the flight Chap_23_X-wind_failure. The initial placement and wind is identical to the crosswind landing you did in Chapter 21. This time, one of your engines will fail (we're not telling which one), and you'll have to make it to the airport and land. As with any single-engine landing, carefully manage your power and drag to make sure you can arrive at the airport.

There are some aspects of multiengine flying that occur only in training and some that we hope only occur in training. You'll start with one we never simulate in real-world training but that you can actually try in the simulator. It's the worst-case scenario: an engine failure just as you rotate for takeoff.

Now for the big time. You've gotten a feel for how engine failures affect your flight, but you've always had altitude to spare when it happened. This time you'll fail an engine right after takeoff and try to make it back to the airport in one piece. The odds are a bit against you by working from an airport that's already a mile above sea level, but it's better you see a tough scenario on the sim than in the real world.We'll stack the deck a bit in your favor the first time by using the lightly loaded Baron. Load the flight Chap_23_Bad_luck_1. You're on the end of Runway 29R, and the winds are light. Get yourself ready for takeoff.Press Shift+4 to open the throttle quadrant and place it where you can see it. You can't set the timing of an engine failure to simulate the worst-case scenario of a failure right at rotation. Instead, position your cursor over the OFF position in the left fuel selector. Advance both throttles with your joystick control, and roll for takeoff.

9)]When you hit V_r of 90 knots, rotate for takeoff, and then click your mouse button to cut off fuel to the left engine (see Figure 23-18).

0)]The next few seconds will determine whether you succeed or crash. Success requires four things: getting the gear up, maintaining blueline, feathering that left propeller, and getting turned away from the rising terrain ahead of you. You'll do these things quickly and in that order. Gear up and pitching for 101 can happen at once, really. Quickly pump the left throttle back halfway and then full forward to make sure you've got the correct engine, then pull that throttle to idle. Then pull the left prop control halfway back, pause for a heartbeat, and then pull it to feather. Turn right into the good engine, and get back to lower terrain and an airport (see Figure 23-19).If you're having trouble, don't fret. It's hard. You can practice a bit at a lower altitude by resetting the flight and changing the airport to KPAE, which is Paine Field in Everett, Wash. There you're only 600 feet MSL. This makes the right-turning tendency worse when the engine fails, but you'll have more power to play with on the good motor. Power is everything with a failure on takeoff. You did remember to lean before takeoff in Colorado, didn't you?Not all takeoffs with a failure are even possible. Load the flight Chap_23_Bad_luck_2. This Baron is at gross weight. Try the same engine failure. Did you make it? What happens if you rotate at 101 instead of 90 and fail the engine then? It's actually possible to fly this flight and land at an airport without crashing, but it's not easy ... and the airport might not be Jeffco. (Yet another difficult takeoff is described in "The Telluride Challenge" at right.)

The Commercial Pilot Practical Test Standards for the multiengine rating includes a demonstration of V_mc. The real-world demo involves just bringing an engine to idle, not actually shutting it down. Recovery is also at the first loss of control. Here, you'll shut the engine down and run right into a V_mc stall for fun. Hey, it's a simulator, right?Load the flight Chap_23_multi_demos. You're back near Seattle, just 3000 feet MSL. You'll want some power to do these well, so you'll leave Colorado for a bit. You're level at 3000 feet on a heading of 330 and a speed of 101. Hold that heading, and shut down the left engine with the fuel selector. Go to full power and full-forward prop on the right engine. Use right rudder and a slight (2-3 degrees) right bank to hold your heading.Now pitch up slowly to get below V_mc of 85 knots. You'll need nearly full right rudder to keep flying straight if you don't put in too much bank. This would be the limit of directional control in a real-world test.Now go further. The FSX Baron will let you get to about 75 knots before it stalls. At this point, you'll see that you can't keep going straight without a significant right bank. That's V_mc.Pull up further if you want and stall to see what happens. The FSX Baron will be forgiving in that you actually can recover by reducing the power and pitching down the nose. The real Baron might start a spin to the left, which is a position you never want to see in a real multiengine airplane.

Reset the flight with Ctrl+;. This time you'll fully feather one engine and look at how various speeds and drag items affect your rate of descent.Shut down the left engine with the fuel selector and feather the prop. Now open the cowl flaps on the right engine and adjust the power so you're maintaining a constant altitude. It should require about 22 inches of MP with full rpm. This is your baseline, and you won't touch the power for the rest of the demo.Any change to your speed or extension of flaps or gear should create a rate of descent. Try flying at V_yse -10 knots, or 91 knots. You'll see an initial climb, but as you hold 91 knots and don't adjust the power, you'll start coming back down about -300 fpm. Now try V_yse +10 knots, or 111 knots. You'll see a big drop at first, but then you'll settle down to a steady rate of descent at about -400 fpm.

1)]Return to 101 knots, and you'll get back to level flight. Drop the gear, and note your descent. Now add approach flaps. Now add full flaps (see Figure 23-20). Now try gear up but full flaps ... you get the idea. Feel free to fill out the following chart as you go:

When you're done, get ready to fly faster and higher than ever before: It's time for your King Air checkout in Chapter 24.

The following are some key points from this chapter:

Here are the FSX lessons and missions to study after reading this chapter:

[Editor's Note: This is the last of three chapters from Microsoft Flight Simulator X for Pilots that AVweb will reprint. If you want to read the whole book, you can purchase it from the AVweb Bookstore.]