A second engine sounds like a spare. Lose one, keep flying on the other. That is the reason many pilots give for wanting a multi-engine rating, and it is only half true. On a typical light twin, losing one engine costs you half your power, and according to the FAA’s Airplane Flying Handbook that reduces climb performance by 80 to 90 percent, sometimes more.

The other half of the problem is worse. The good engine is now pulling from one side of the aircraft only, and it wants to swing the nose towards the dead one. Below a certain airspeed, the rudder simply cannot stop it. That airspeed has a name, a red line on the airspeed indicator and a starring role in every multi-engine checkride: V_MC.

What a multi-engine rating actually is

In FAA terms, flying a light twin requires a multiengine class rating on top of a pilot certificate. Much of the training revolves around one scenario: what to do, and how fast, when an engine quits. Most schools teach it in purpose-built trainers such as the Piper PA-44 Seminole, which Wikipedia notes is used primarily for multi-engine flight training, or the larger Piper PA-34 Seneca.

On the practical test, the candidate has to fly an “engine inoperative, loss of directional control demonstration”, which the FAA handbook says is often referred to as a V_MC demonstration. Understanding what V_MC means, and what moves it, is the core of the rating.

The handbook lists three major accident factors when flying on one engine at low altitude and low airspeed, such as just after take-off: loss of directional control, loss of performance and loss of flying speed. It adds that all three have equal potential to be lethal.

V_MC: the speed where the rudder runs out

V_MC is the minimum control speed with the critical engine inoperative. In certification, it is the sea-level calibrated airspeed at which a test pilot, after the critical engine is suddenly made inoperative, can keep control and then hold straight flight at the same speed with no more than 5° of bank.

Under the small-aeroplane certification rules described in the handbook, the pilot must stop the resulting turn within 20° of the original heading using maximum rudder and no more than 5° of bank. Crucially, there is no requirement that the aircraft can climb at V_MC. It is a control speed, not a performance speed.

That is why the airspeed indicator of a light twin carries two extra marks. The red radial line is V_MC. The blue radial line is V_YSE, the best rate-of-climb speed with one engine inoperative, the speed you want if an engine fails after take-off. There is also V_SSE, the safe intentional one-engine-inoperative speed, the minimum at which an instructor may deliberately shut down or simulate failure of the critical engine.

Why one engine is more critical than the other

On most twins, both propellers turn clockwise as seen from the pilot’s seat. At positive angles of attack, the descending blade of each propeller produces more thrust than the ascending blade, the familiar P-factor. On the right engine, that descending blade sits farther from the aircraft’s centre of gravity, so it has a longer lever arm.

If the left engine fails, the right engine’s thrust acts on that longer arm and produces the bigger yawing moment. That makes the left engine the critical engine: the one whose failure has the most adverse effect on directional control. Published V_MC is measured with that engine out.

Flight instructor Martin Pauly explains V_MC and the minimum control speed demonstration in multi-engine training.

Designers have a neat fix. The Seminole and Seneca both use a right-hand engine that turns the opposite way, a counter-rotating arrangement that Wikipedia says eliminates the critical engine. With counter-rotating propellers, the FAA handbook notes, the asymmetric thrust is the same whichever engine fails. The P-38 Lightning fighter of the Second World War also had counter-rotating propellers, turning outward at the top of the arc.

The number on the dial is not the number in the sky

The red line is a single figure, but the handbook stresses that real V_MC changes with conditions. The published value is determined under a precise, deliberately unfavourable set of conditions: maximum available take-off power on the operating engine, the critical engine’s propeller windmilling in the take-off position unless an autofeather system is fitted, the most unfavourable weight and the aft-most centre of gravity, landing gear up, flaps and cowl flaps in the take-off position, trimmed for take-off, out of ground effect and with up to 5° of bank towards the operating engine.

Each of those factors moves V_MC. More power on the good engine raises it. A windmilling propeller adds drag on the dead side and raises it; feathering the propeller removes most of that drag. An aft centre of gravity shortens the rudder’s lever arm and raises it. Retracting the gear raises it, because extended gear aids directional stability. Counter-intuitively, V_MC also increases as weight is reduced.

Bank angle is the most sensitive of all. Banking towards the good engine lets a component of lift help the rudder. According to the handbook, tests have shown V_MC may rise by more than 3 knots for every degree of bank less than 5°. Fly wings-level with a failed engine and the real minimum control speed can sit well above the red line.

When V_MC meets the stall

Altitude adds a twist. With normally aspirated engines, V_MC is highest at take-off power at sea level and falls as the aircraft climbs, because the good engine produces less power in thinner air. Turbocharged engines hold take-off power, and so V_MC, up to their critical altitude. The indicated stall speed, meanwhile, stays roughly where it is. At some altitude the two meet.

That is the dangerous corner of the envelope. The handbook warns that if the stall arrives at or before V_MC, the departure from controlled flight may be sudden, with strong yaw and roll towards inverted and a spin entry. No multi-engine aeroplane is approved for spins, and their recovery characteristics are generally very poor.

Hence the rules for training. The handbook says pilots should never demonstrate V_MC with an engine cut from high power and never intentionally fail an engine below V_SSE. The demonstration is flown high, planned so it can be completed at least 3,000 feet above ground, and at the first sign of uncontrollable yaw or a stall, the recovery is to retard the operating engine’s throttle enough to stop the yaw while lowering the nose.

Flying the failure: zero sideslip

Once control is assured, the goal is performance. A twin flying on one engine climbs best at zero sideslip, which for a typical light twin the handbook puts at about 2° of bank towards the operating engine, with the slip ball displaced one-third to one-half towards the operating engine. More bank than that adds drag and eats the little climb you have left.

The AOPA Air Safety Institute’s Taming the Twin introduction to multiengine aeroplanes.

The airline world solved the same problem with bigger numbers and stricter rules, as Afterburner explained in Engines Turn or Passengers Swim. Burt Rutan attacked it from another angle with an aircraft deliberately built asymmetric: The Rutan Boomerang: Built Lopsided on Purpose.

The multi-engine rating is not hard because twins are hard to fly. With both engines running, they are often easier. It is hard because the one moment that matters arrives without warning, close to the ground, at low speed. Keep the speed above the red line, fly the blue line, and a second engine really does become what it promises to be.

A V_MC demonstration flown in a Diamond DA42 twin trainer.

Sources: FAA Airplane Flying Handbook (Transition to Multiengine Airplanes), Wikipedia (Critical engine; Piper PA-44 Seminole; Piper PA-34 Seneca), AOPA Air Safety Institute