Whoa, nellie, the procedure I’ve been taught for recovery from wing stalls in multi-engine airplanes is all wrong!

In single-engine aircraft, the normal procedure for stall recovery is bumping the nose down, then immediately applying full engine power. This minimizes the amount of time the aircraft is flying too slowly for the wings to keep it up. I recall being taught to do the same thing on the Aztec, and that’s how I’ve done practice stalls ever since.

Reading a set of reports made me rethink the issue. Consider what happens if, early during the stall recovery, one of the engines quits. The airspeed is still low, one engine producing full power, the other is windmilling: a recipe for loss of control due to asymmetric thrust, possibly resulting in an unrecoverable spin.

The proper practice stall recovery procedure in twins makes airspeed control far more important than in singles. One still pushes the nose over first, but instead of applying power right away, one has to let the aircraft dive long enough to exceed V_mc (80 mph in the Aztec). Only after that point is it safe to open up both throttles and recover from the dive / loss of altitude.

If one encounters an unintentional stall at a low altitude, then one has to decide instantly which is the worse evil. I expect pilots to accept the risk of an engine-failure-induced spin in this situation and add power as soon as possible to arrest the stall/descent. This is another of those interesting “no-win” corners of the flight condition envelope where operating twins involves instant decision making.

This reminds me of a mistake I’ve often committed while practicing one-engine-inoperative maneuvers in the airplane: lack of sufficient airspeed. There is an ideal speed V_yse (100 mph) marked right on the gauge for climbing with one engine, which is about 70% of the typical one-engine cruise speed. Even when there is no need to climb, for some reason I sometimes slow down to that speed. Worse, I accomplish that by reducing power on the good engine, wasting a perfectly good energy resource in a time of (simulated or real) emergency. It’s as if I imagine that V_yse was the proper speed for all single-engine operations.

It turns out that the Aztec climbs nicely even at a speed well above V_yse, so even for climbs this low a speed is not necessary. Going even slower can cause so much drag (because of the high angle of attack) that the airplane can’t climb at all. So really V_yse should better be considered a minimum speed, with a faster speed preferred. If this requires higher power on the good engine, so be it: pour it on.

Learning about flying seems neverending.