When I got my instrument rating in 1967 and my CFII in 1971, skill at flying needle, ball, and airspeed received a lot of emphasis. And with good reason: the other gyro instruments (attitude and heading) were vacuum-driven, and the vacuum systems had the nasty habit of failing without warning. Standby vacuum systems didn’t yet exist. So partial-panel proficiency was considered a crucial survival skill for an instrument pilot.
The emphasis on partial-panel flying surged in the mid-60s. About that time, the manufacturers of piston-powered aircraft, in their collective infinite wisdom, decided to change en-masse from all-metal oil-lubricated “wet” vacuum pumps to state-of-the-art “dry” pumps which used rotors and vanes made of graphite and which therefore required no lubrication.
The advent of dry air pumps was not exactly a boon to instrument pilots. While the old-style wet pumps just gradually get weaker as they wear out, the new dry pumps work at 100% efficiency until poof! they self-destruct into a cloud of carbon dust without even a moment’s forewarning. Furthermore, the lifespan of a dry air pump is quite unpredictable: the pump might have 10 hours or 1000 hours on it when it decides to fail suddenly and catastrophically.
Backup Vacuum SystemsThe invention of air pumps that fail unpredictably and without warning created an opportunity for firms who developed and STC’d backup vacuum systems for single-engine airplanes. These first became popular in the ’80s. There are basically five kinds:
Dual full-time engine-driven dry air pumps, such as are factory-installed on late-model Cessna P210s. Both pumps operate all the time and share the load. This means, unfortunately, that both pumps wear out all the time. If one pump fails, the other pump takes the full load…which might just be sufficient to make it fail, too. Twin-engine airplanes have essentially the same system (except that one pump is driven by each engine), and share the same vulnerability.
Dual engine-driven dry air pumps, a full-time primary pump plus a standby backup pump with an electrically-controlled clutch, such as the RAPCO system factory-installed in late-model Beech Bonanzas. Failure of the primary pump is supposed to bring the secondary pump on-line. Unfortunately, there’s no good way for the pilot to test whether the clutch and the standby pump is working properly until the primary pump fails. Then, either the standy pump takes over…or it doesn’t.
Dual dry air pumps, one full-time engine-driven primary pump plus a standby backup pump driven by an electric motor controlled by a switch in the cockpit. Several STC-holders offer such systems, and they are the best of the lot. They are also the most expensive. They can be tested before engine-start to make sure the standby system works. And the standby pump doesn’t run (or wear) except when it’s actually needed.
One full-time engine-driven primary pump plus a secondary system that uses engine intake manifold vacuum as a backup vacuum source. This system is manufactured by Precise Flight and is very popular because it is far less expensive than other backup systems. Unfortunately, it has several drawbacks. If the primary pump fails, the system only provides the needed 4-5″ of vacuum to the gyro instruments when the engine is being operated at partial throttle. This means that it won’t work at high altitudes (when full-throttle is required to hold altitude), nor during other operations that require full-throttle…such as a missed approach! Furthermore, this system won’t work on a turbocharged engine.
One full-time engine-driven primary pump plus an externally mounted venturi providing a backup vacuum source. The venturi may be heated electrically to prevent it from icing up. Because the venturi is ugly and creates drag, it hasn’t proven very popular among aircraft owners.
Lots of single-engine aircraft still have no backup for their one engine-driven dry vacuum pump that can fail suddenly without warning. And lots more have standby vacuum systems that may or may not work when the primary pump fails (surprise!), or may cease to work during full-throttle operation. A chafed hose or loose hose clamp can cause a total vacuum system failure even in an airplane with a fully-functional standby system. And the vacuum-driven gyro instruments themselves fail quite often (due to wear and contamination) even when the vacuum system itself is working okay.
To my way of thinking, instrument flight by reference to needle, ball, and airspeed is as important a survival skill for the IFR pilot as it ever was. And because we don’t use it in our day-to-day flying, partial-panel flying is a skill with a very short half-life. If you don’t practice it at least twice a year, it won’t be there when you need it most.
Inside the Turn-and-Slip IndicatorThe gyroscopic turn-and-slip indicator (sometimes called turn-and-bank) differs from the other gyro instruments (attitude and heading indicators) in several important ways:
It is electrically-driven, not air-driven, so it continues to function even when the vacuum system fails.
It is a completely sealed unit, which makes it much less likely to fail than an air-driven gyro. (Bearing contamination is perhaps the primary reason that air-driven gyros fail.)
Its internal construction is far less complex than other gyro instruments, which again makes it much less likely to fail. The turn-and-slip indicator contains a massive gyro wheel mounted with its axis-of-rotation parallel to the lateral (left-right) axis of the airplane. It spins up and away from the pilot and, in modern designs, is driven by a brushless DC motor with permanently-lubricated sealed bearings.
When the aircraft yaws, the gyro assembly is forced to yaw along with it. Gyroscopic precession causes the gyro to tilt left or right against the force of the centering spring. The higher the rate of yaw, the greater the precession force against the spring and the farther the gyro tilts.
Because the gyro wheel rotates up and away from the pilot, the gyro assembly actually tilts to the right when the aircraft yaws left, and to the left when the aircraft yaws right. Consequently, the gyro gimbal is linked to the turn needle through a reversing linkage that causes the needle to deflect in the direction of the yaw.
The instrument also contains a dashpot in order to slow down the movement of the gimbal and eliminate minor wiggles and oscillations of the turn needle.
The face of the instrument is inscribed with two peaked marks referred to as “doghouses” that indicate standard-rate 3-degree-per-second turns to the right or left.
The “slip” part of the turn-and-slip instrument is simply a heavy ball in a curved glass tube filled with kerosene to dampen its movements, with some scribed marks to show when the ball is precisely centered. Carpenters used identical technology for centuries before the Wright Brothers flew.
Inside the Turn CoordinatorIn the early ’70s, the venerable turn-and-slip instrument was redesigned, and the newer version was called a “turn coordinator”. The turn coordinator is constructed in identical fashion to the older turn-and-slip instrument, with three significant differences:
The gyro gimbal axis is tilted about 30 degrees from horizontal (fore pivot high, aft pivot low). This causes the gyro to react to both rate-of-yaw and rate-of-bank.
The dashpot is replaced by a viscous dampener, which causes the turn coordinator to react less to turbulence than the traditional turn-and-slip instrument.
The traditional turn needle is replaced by a miniature airplane symbol that tilts its wing to indicate the direction and rate of turn. Hash marks replace the traditional “doghouses” to indicate standard-rate turns to the left or right.
Nowadays, however, many airplanes are equipped with turn coordinators that are not connected to the autopilot. Some pilots find it easier to hand-fly partial-panel using a turn coordinator. Others prefer the traditional turn-and-slip instrument with its yaw-only gyro and its less-damped needle.(See "Turn-and-Slip vs. Turn Coordinator.")
When the Vacuum Gauge Reads ZeroI’m something of a nut on recurrent training. I practice partial-panel flying a lot. I’ve been doing so for close to 30 years now. I still find that it’s damned hard to do well, and very easy to screw up big-time.
Don’t forget that when your air-driven gyro instruments fail for real, there won’t be a CFI slapping no-peekies over them. The heading indicator will simply start to precess at an increasing rate, and the attitude indicator will start developing a gradual case of “the leans” long before it starts twitching spastically.
If your autopilot uses these instruments as its attitude and heading sensors (which it probably does if you have more than just a one-axis wing-leveler), it will keep trying to fly the failing gyros right into a graveyard spiral. Plenty of pilots have died in exactly this fashion. If you don’t want to be one of them, you need to keep up your instrument cross-check anytime “George” is engaged. If the instruments seem to be arguing with one another, you need to decide which ones are lying. And if your autopilot depends on those, you need to disengage it quickly and start hand-flying.
On the other hand, if you have a simple wing-leveler autopilot that uses a turn coordinator as its only gyro input, you’ve really lucked out. For heaven’s sake leave the thing on, because you’ll need all the help you can get!
When you find yourself flying in needle-ball-airspeed mode, one of the first things you’ll want to do is to cover up the failed gyro instruments. If you don’t, you’ll find it almost impossible to ignore them. Post-it notes work great for this…you can even use a couple of business cards in a pinch.
What makes partial-panel flying so blasted difficult is that (1) it is awfully easy to over-control, and (2) it is almost impossible to recover from an unusual attitude. So: you need to limit yourself to very subtle control inputs, and you need to avoid getting into an unusual attitude at all costs.
To keep control inputs subtle, I find it useful to pretty much keep my mitts off the control yoke. Instead, I make pitch inputs with trim and roll inputs with rudder. It’s a funny way to fly, but it makes it much harder to overcontrol.
We practice both wet compass turns and timed turns when we prepare for an instrument rating ride, but both involve more pilot workload than is reasonable when actually flying in the system in IMC with a single pilot. Therefore, if you lose your gyroscopic heading indicator, I suggest you immediately explain your problem to ATC and request “no-gyro vectors” to the nearest suitable destination. Let the controller deal with your navigation problem so you can give your full attention to your task of basic partial-panel instrument flying.
Assuming you have the airplane more-or-less under control in straight-and-level cruise flight, your next priority is to get either into VMC or on the ground promptly. If the ceilings are high enough or the tops low enough, ask for a no-gyro descent or climb to get out of IMC.
Otherwise, you’ll have to make a partial-panel instrument approach. And, by the way, this is perfectly acceptable grounds for declaring an emergency and asking for priority handling. I know that *I* would declare.
As far as I’m concerned, the *only* acceptable instrument approach to execute partial-panel is a precision approach (full ILS or PAR). The reason for this is that making major changes in aircraft configuration can be truly treacherous while flying partial-panel, and non-precision approaches generally require at least several configuration changes (descent, level-off, descent, level-off, etc.), each one of which is an invitation to disaster.
So find a nearby airport with an ILS and with weather well above ILS minimums. Tell ATC you need to make the ILS there, and request no-gyro vectors to the final approach course. Explain to the controller that the further out you can intercept the localizer and the higher you can intercept the glideslope, the better chance you have of making a successful approach. The controller is likely to be cooperative if you explain your needs to him…particularly if you have also declared an emergency.
Reduce to approach speed and drop the gear and flaps as early and high as you can. It is during these configuration changes that you are most likely to “lose it”, and 1500′ AGL is not the best place to do this stuff. Use pitch trim to achieve the desired approach speed. Hands off the yoke, lest you be tempted to overcontrol.
For the same reason, make awfully darned sure that you won’t have to miss the approach. Executing a partial-panel missed approach in IMC is a good way to get yourself killed. Major configuration change at very low altitude. Not a good idea at all.
Once ATC has no-gyro vectored you to the point where your localizer needle comes alive, don’t try to fly the localizer in the usual fashion. Instead, simply concentrate on one thing: stopping the needle from moving. If you see the localizer needle moving left, press left rudder to achieve a half-standard-rate turn to the left until the needle stops moving, then level the wings. Likewise if you see the needle moving right. Don’t worry about trying to center the needle…just work to keep it stationary. Even if you break out with the localizer needle several dots off-center, you’ll still be in a fine position to land.
When the glideslope needle comes alive, treat it the same way. Don’t worry about trying to keep it centered. Just concentrate on stopping its movement. If you see the needle headed downward, reduce power a touch. Upward, increase power a touch. If the needle is stationary, leave the power alone. Your airspeed may vary above or below the normal approach speed. If it’s within 10 knots of nominal, don’t sweat it. If it goes beyond that, adjust the pitch trim a little. Hands off the yoke.
Your scan is focussed on the ILS head and the turn-and-slip or turn coordinator, with an occasional glance at the airspeed. Ignore the compass, VSI, and altimeter.
If you’ve never tried this “stop-the-needles” approach to flying a partial-panel ILS, go practice it with your CFII or safety pilot. It’ll feel incredibly weird at first…but you’ll be amazed at how well it works. Once you get the hang of it, you might just find that you make smoother ILS’s partial panel than you do with all the gyros working!