Temperature and Angle-of-Attack Measurement

OAT, SAT and TAT

Three names, two temperatures:

Outside air temperature (OAT) — the actual temperature of the undisturbed air the aircraft flies through; measured properly it is the same thing as the static air temperature (SAT), the term the instruments and computers prefer.

Total air temperature (TAT) — what a probe on a fast aircraft actually feels: air arriving at the probe is compressed and brought nearly to rest, and compression heats it, so the probe senses the static temperature plus a heating due to speed.

The relationship mirrors the pressure family deliberately: as total pressure is static plus dynamic, total temperature is static plus the ram rise.

Ram rise, and converting TAT to SAT

The ram rise is the difference TAT − SAT: the temperature added by bringing the moving air to rest against the probe. For a given air temperature it depends only on the Mach number, growing with its square, negligible for a light aircraft and around 25 to 30 degrees for a jet at cruise. With temperatures in kelvin the conversion is one line, written for a probe that captures the whole ram rise (called full recovery, which modern TAT probes very nearly achieve):

Probe locations and errors

Two thermometer designs:

Direct-reading thermometer — a bimetallic element, two bonded metals that curl as they heat unequally, in a small housing read where it is mounted: it suits light aircraft, through the windscreen frame or the cabin's fresh-air inlet, where the pilot can see its dial.

Remote-reading probe — suits everything larger: the TAT probe above, mounted on the forward fuselage clear of the boundary layer, its electrical signal driven to displays and computers.

Two of the pitot-static lesson's three error families return here:

Position error — a probe too close to the skin sits in air the fuselage has already slowed and warmed (the boundary layer), or worse, behind an engine or an air outlet; the remedy is location.

Instrument error — the sensing element's own tolerances and ageing; the remedy is build quality and calibration.

Probe heating and its side effect

Temperature probes ice up like pitot tubes, and an iced scoop reads nonsense, so the probes are electrically heated. The purpose is the same as pitot heat: keep the sensor clear of ice. But heating a thermometer has an obvious side effect: some heater warmth reaches the sensing element, and the probe over-reads — unless the design compensates automatically, which modern probes do by geometry (the bleed flow carries heater warmth away from the element) and by correction in the computer. The syllabus point to keep: heat on means a slightly high reading on an uncompensated probe, and the direction of the error is always warm.

The angle-of-attack sensors

Angle of attack, the angle between the oncoming flow and the wing, is the quantity that actually decides the stall, and the aircraft measures it with one of two sensor types:

Vane detector — a small weathercock on the fuselage side: a pivoted vane streams with the local airflow, and the shaft's angle, measured electrically, is the angle of attack.

Null-seeking (slotted) probe — a rotatable probe with two slots, one each side of its centreline. Meeting the flow at an angle, one slot feels more pressure than the other; the difference drives a motor that rotates the probe until the two pressures are equal (the null), at which point the probe is aligned with the flow and its rotation angle is the measurement.

Both types are protected against ice by electrical heating, in the vane's case heating both the vane and its case, because a frozen AoA sensor is worse than a dead one, as the closing section shows.

Who uses the angle of attack

The AoA signal fans out to the systems that live or die by the margin to the stall:

the air-data computer, which uses it among other things to refine its corrections.
the stall warning system, the stick shaker and warnings whose aerodynamic basis Principles of Flight built.
the flight-envelope protection of fly-by-wire aircraft, which limits the angle of attack the pilot can command in the first place.

Reading the AoA displays

Where AoA is displayed at all, two presentations dominate. Light arrays: a short ladder of green, amber and red lights beside the glareshield (the anti-glare shelf above the instrument panel), read like a traffic signal:

green — an approach-appropriate angle.
amber — a margin shrinking toward the stall.
red — an angle demanding immediate reduction.

Some arrays use the pattern (which lights, how many) to show fast/slow trend for the approach. Gauges with a numerical scale: a dial or tape showing the angle itself, often normalised (0 to 1 from zero-lift to stall) rather than in raw degrees, with the working band and the stall region marked. Both say the same thing in different accents: how much wing is left.

When the AoA data goes wrong

The dangerous failure is not a dead sensor but a plausible liar: a vane frozen in a fixed position keeps sending a steady, believable angle regardless of what the wing is doing. The pilot's instruments then disagree quietly: stall warnings may fire in normal flight (frozen at a high angle) or stay silent while a real stall approaches (frozen at a low one), and airspeed and attitude no longer tell the same story as the AoA-driven cues. The defence is the cross-checking habit this course keeps returning to: an AoA-driven warning that contradicts healthy airspeed, attitude and thrust readings is treated as a sensor question first.

With flight-envelope protection the stakes rise, because a protection system acts on the measurement. Fed a falsely high angle of attack, the protection can conclude the aircraft is stalling and command nose-down pitch or block nose-up inputs while the wing is actually flying normally: the aircraft becomes hard to control not because its aerodynamics failed but because its protections are fighting a phantom. Fed a falsely low angle, the protections stand by while the real margin vanishes. This is why AoA sensors are multiplied and voted on fly-by-wire aircraft, why sensor disagreement downgrades the control laws to give authority back to the pilot, and why the crew drills for protections misbehaving include switching the automation out of the loop.