ATPL 022 Instrumentation common mistakes
ATPL 022 Instrumentation · Understand the traps and how to avoid them.
When a question describes a static port blockage or a pitot tube blockage, the candidate applies the same reasoning to both, concluding that 'the ASI reads high in a climb' regardless of which port is blocked, or assuming the altimeter and VSI are also affected by a pitot blockage.
Pitot and static blockages are taught together, so the two get merged into one 'blocked probe' memory item. The candidate never builds a clean mental model of which instrument uses which pressure source: the ASI compares pitot (total) pressure with static pressure, while the altimeter and VSI use static pressure only. Without that separation, the direction of the error in a climb or descent cannot be derived, only guessed.
What to do instead
Before answering, write down the two pressure inputs for each instrument. Then reason step by step: with a blocked static port, static pressure is trapped at the value at the moment of blockage, so in a climb the ASI over-reads (trapped static is higher than actual) and the altimeter under-reads; in a descent the reverse. With a blocked pitot (drain open), the ASI behaves like an altimeter in reverse. Practise deriving each case from the pressure equations rather than recalling a slogan.
The candidate treats the turn indicator and the turn coordinator as interchangeable, or states that the slip/skid ball responds to yaw rate rather than to lateral acceleration, and misidentifies which gyro axis drives each display.
Both instruments look similar on the panel and both are described as 'rate' instruments, so the candidate skips the distinction. The deeper cause is not having linked each display to its gimbal arrangement: the turn indicator uses a rate gyro whose spin axis is horizontal and lateral, sensing yaw rate, while the turn coordinator cants the gyro so it senses both yaw and roll rate, giving lead in the roll-in. The ball is an inclinometer sensing the resultant of gravity and centrifugal acceleration, not a yaw sensor.
What to do instead
Draw the two gimbals side by side and label the sensing axis and the presentation (needle vs miniature aircraft). State explicitly that the ball indicates lateral acceleration (slip/skid), which is why it can be central in a steady turn and off-centre in a skidding turn at constant heading rate. Test yourself by predicting the indication in a coordinated level turn, a slipping turn and a skidding turn.
On questions about gyroscopic drift, the candidate attributes apparent wander to real wander, or forgets that apparent drift depends on latitude and heading, and answers as if drift were a fixed hourly figure independent of position.
Textbook values such as '15 degrees per hour' are memorised without the underlying relation to Earth rate, so the candidate cannot apply the latitude and heading dependence. Real wander (bearing friction, imbalance, manufacturing imperfection) and apparent wander (Earth's rotation seen in the gyro frame) are both called 'drift', and the candidate does not keep the two causes separate.
What to do instead
Anchor on the rule that apparent drift is zero at the equator, maximum at the poles, and depends on the sine of latitude and on the gyro's heading relative to the meridian. Work a numerical example: at 60 degrees north, the horizontal component of Earth rate is about 7.5 degrees per hour, so a free gyro's apparent wander is largest when its spin axis is horizontal. Then list real-wander causes separately and note that they are unpredictable and not latitude-dependent.
The candidate confuses variation and deviation, or applies the wrong sign when converting between true, magnetic and compass headings, and fails to check the direction of the correction (east is least, west is best) against a compass swing card.
Variation and deviation are both angular errors between norths, and the mnemonic 'east is least' is remembered without being tied to the actual arithmetic. Candidates also forget that deviation is aircraft-specific and heading-dependent, so it must be read from the deviation card at the relevant compass heading, not applied as a single constant.
What to do instead
Build a three-row ladder (true, magnetic, compass) and always write the correction with its sign and direction. Practise with a deviation card: read the deviation for the compass heading in use, apply it to get magnetic, then apply variation to get true. Do five conversions in each direction until the sign convention is automatic, and state the rule 'from compass to true, add east' explicitly each time.
When asked what happens after a total electrical failure or a generator trip, the candidate assumes all instruments are lost, or conversely assumes standby instruments are unaffected without checking whether they are independent of the main bus.
Panel layouts are learned as a picture rather than as a power architecture. The candidate does not know which instruments are pitot-static (no electrical supply needed), which are electrically driven gyros, which are fed from the essential or standby bus, and which have their own battery. Without the power-source map, any failure question becomes guesswork.
What to do instead
Make a table with columns for instrument, sensing principle, power source and backup. Mark the pitot-static trio as power-independent, the attitude and heading gyros as electrically driven with standby or battery backup, and the electronic displays as dependent on their symbol generators and buses. Then rehearse the consequence chain for loss of the main bus, loss of a single generator and loss of all AC.
The candidate answers questions about air data computer outputs or electronic display redundancy by describing the sensors rather than the processing path, and cannot say what the crew sees when one air data computer disagrees with another.
The ADC is treated as a black box that 'computes airspeed', so the candidate never learns that it converts raw pressures and temperatures into corrected outputs (CAS, Mach, TAS, altitude, vertical speed, temperature) and distributes them to displays, the autopilot and the flight director. Comparison and voting logic between redundant computers is therefore invisible, and disagreement scenarios are answered by intuition.
What to do instead
Trace one parameter end to end: pitot and static pressure plus total air temperature enter the ADC, are corrected for position error and instrument error, and leave as digital outputs on the data bus to specific users. Learn that modern systems compare sources and flag a disagreeing source rather than silently averaging, and that the crew response is to cross-check with standby instruments and isolate the faulty source.
On autopilot and autothrust questions, the candidate mixes up modes and their arming/capture logic, for example stating that selecting a vertical speed mode captures altitude, or that autothrust controls pitch.
Mode annunciators are memorised as a list of names without the control-loop structure behind them. The candidate does not separate the pitch channel (which controls attitude, vertical speed or altitude) from the thrust channel (which controls speed or thrust), nor the armed state from the engaged state, so mode transitions and their annunciation cannot be predicted.
What to do instead
Draw two columns, pitch and thrust, and place each mode under its channel with its captured and armed states. Note that vertical speed mode must be disconnected or transitioned before altitude capture, and that autothrust commands thrust to hold a target speed or a fixed thrust rating, never pitch. Then walk through a climb from takeoff to cruise, naming the active and armed modes at each stage.
The candidate confuses the different magnetic compass errors (turning, acceleration, northerly turning error, oscillation) and applies the wrong one to a given manoeuvre, or forgets that the direct-reading compass is only reliable in steady, level flight.
The errors are taught as a list of names, and the candidate does not connect each error to the physical cause: the compass card's centre of gravity below the pivot, and the dip of the magnetic field lines. Without the mechanism, the direction of the error in a turn through north or south, or during acceleration and deceleration on easterly and westerly headings, cannot be worked out.
What to do instead
Learn the mechanism first: the card hangs nose-down in the northern hemisphere, so in a turn through north it lags and through south it leads, and acceleration on an east or west heading causes a false turn indication. Then use the rule that the compass is read only in steady flight, and that the standby or gyro heading is used during turns. Practise predicting the indication for turns through each cardinal heading.
When a question describes an inertial reference system or GNSS problem, the candidate cannot distinguish a gyro or accelerometer fault from an alignment, power-interruption or database problem, and treats all IRS degradation as 'drift'.
INS/IRS is learned as a single navigation box rather than as a strapdown or gimballed platform with a defined alignment procedure and defined failure indications. The candidate does not know the difference between the alignment modes, the effect of aircraft movement during alignment, and the fact that attitude and heading can remain valid while position degrades. GNSS questions add a second layer (RAIM, integrity monitoring, database currency) that is not linked to the inertial side.
What to do instead
Separate the failure categories: alignment failure (movement during alignment, insufficient time, wrong position entry), power interruption (full realignment required, attitude may be lost), sensor failure (flagged by the system, with reversion to other sources) and long-term drift (position error growth, corrected by radio or GNSS updating). For GNSS, learn integrity monitoring, the effect of satellite geometry on accuracy, and the requirement to check database currency.
On data bus and computer questions, the candidate describes the bus as a set of point-to-point wires and cannot say what happens when one user or one line fails, nor why the bus architecture changes the failure behaviour compared with older discrete wiring.
The architecture is abstract and rarely seen, so the candidate memorises acronyms without the topology. The key property, that a serial bus carries addressed digital messages shared by many users, is not understood, so questions about a single failed line, a failed user or a bus contention cannot be reasoned through.
What to do instead
Learn the essentials of the specific architecture: a twisted-pair or similar medium, serial transmission of addressed messages, and each user taking only the messages it needs. Then reason about failures: one failed line can take down the bus for all users if no redundancy exists, a failed user can be isolated, and redundancy is provided by duplicate buses. Compare this with discrete point-to-point wiring to see why weight and fault isolation improve.