Pilot Controls — Sidesticks, Pedals and Trim

Flight Control Fundamentals drew the line that the whole chapter runs along: the surfaces are electrically-controlled and hydraulically-actuated. This article enlarges the single most upstream box on the control side — the layer of hardware between the pilot's hand and foot and the electrical signal that reaches the computers. That layer is the sidesticks, the rudder pedals, the speedbrake lever, the pitch trim wheel, the rudder trim knob, and the transducers that turn each of their positions into an electrical order. In the maintenance source they live in one chapter, 27-92 Control Inputs and Power Supply, because they all do the same job: translate how much the pilot wants into a voltage and feed it to the five flight control computers.

The whole layer is built on one engineering idea: almost everything here is a pure electrical signal — uncoupled, no feedback, and redundantly split so no single break is fatal. There is exactly one exception, and it is the reason the chapter keeps coming back to it: the manual pitch trim wheel, the only mechanical tail on the aeroplane.

[!warning]- The two sidesticks do not talk to each other, and your hand feels a spring — not the aeroplane.

The sidesticks are not mechanically linked and receive no feedback from the surfaces. Move yours and the other pilot's stick does not move; pull hard and the force you feel is the artificial-feel spring, not the aerodynamic load on the elevator. The one device in this layer with a mechanical run to a surface is the pitch trim wheel to the stabiliser. Carry the "yoke" mental model — sticks that back-drive each other, controls that load up with airspeed — into this chapter and the dual-input, priority, and trim logic will all read wrong.

1. The control-input layer — what this article covers

Everything here feeds the computers described in EFCS Computer Architecture, and the surfaces they drive are deepened in the surface articles. This piece stays strictly on the cockpit input devices themselves — what they are, how they turn a movement into an electrical order, and how they are powered. It deliberately stops at the edge of:

• the dual-input summing / priority / takeover logic of the two sticks — see Sidestick Priority Logic;
• the spoiler allocation and ground-spoiler behaviour behind the speedbrake lever — see Spoilers and Ground Spoilers;
• the THS body itself (screwjack, motors, failures) — see Trimmable Horizontal Stabiliser;
• the rudder servos, travel limit and yaw damper — see Rudder and Yaw;
• the ECAM/PFD indications of these controls — see Controls and Indications.

When those topics surface here they are only named, with the deep dive left to the article that owns them.

2. Architecture — from hand and foot to electrical signal

2.1 Inside one sidestick (read from the AMM assembly figure)

The diagram below is traced from the AMM Side Stick Assembly and Side Stick Transducer Unit figures. It pulls one stick apart: hand grip → universal joint → two independent axis shafts → each with its own artificial-feel spring, damper, and transducer. Read it and you understand why the "feel" is manufactured and why no single break loses a whole axis.

                       ┌──────────────┐
                       │  HAND GRIP   │  Radio PTT switch
                       │              │  sidestick pb (AP disconnect / priority)
                       └──────┬───────┘
                              │
                       universal joint     splits the 2-axis grip motion
                   ┌──────────┴──────────┐
            ┌──────▼──────┐       ┌───────▼──────┐
            │ PITCH axis  │       │  ROLL axis   │  each axis = a "duplicate" shaft
            │ duplicate   │       │  duplicate   │
            │   shaft     │       │   shaft      │
            └──┬───┬───┬──┘       └──┬───┬───┬───┘
               │   │   │             │   │   │
          feel │   │ damper     feel │   │ damper
        springs▼   ▼            elem.▼   ▼
            ┌──────────────┐     ┌──────────────┐
            │ TRANSDUCER   │     │ TRANSDUCER   │
            │ UNIT (pitch) │     │ UNIT (roll)  │
            │ 4 groups × 3 │     │ 4 groups × 3 │  middle-point potentiometers
            │ potentiom.   │     │ potentiom.   │  driven by the duplicate gear
            └──────┬───────┘     └──────┬───────┘
                   │ electrical signals  │
                   └──────────┬──────────┘
                              ▼
            ┌─────────────────────────────────────┐
            │  FCPC 1 / 2 / 3 (PRIM)  +  FCSC 1/2  │  one grip position to all 5
            └─────────────────────────────────────┘

   Also on the assembly: the neutral-locking solenoid (12CA1 / 12CA2)
   └ AP engaged: the FMGEC energises the solenoid → two rollers seat in V-cams
     → the grip is held at neutral with a raised force threshold

Four things to read off it:

1. One grip, two independent axes. Pitch and roll are separate mechanical branches, each with its own feel springs, damper and transducer. A roll input does not move the pitch hardware.
2. "Duplicate" is the keyword. Each axis shaft is duplicated, and each transducer unit holds four groups of three middle-point potentiometers driven by that duplicate mechanism. This is the physical basis for the single-failure tolerance in §3.
3. The feel is manufactured. The springs give you the centring force, the damper gives you the damping — the stick takes no feedback from the surfaces, so the force in your hand is pure spring characteristic, unrelated to the aerodynamic load of the moment.
4. One stick signals all five computers. Each grip position is broadcast to 3 PRIM + 2 SEC; which one is master is decided by the logic in 01.

2.2 The control-input devices at a glance

3. The sidestick — springloaded, uncoupled, no feedback

The starting point is the most counter-intuitive sentence in the chapter. Per FCOM DSC-27-20-30:

The sidesticks are springloaded to neutral. They are not mechanically linked, and do not receive feedback from the flight control surfaces. When the autopilot is engaged, the sidesticks are locked in the neutral position.

That is the whole sidestick philosophy in three clauses. Take them in turn.

Springloaded to neutral. Let go and the stick self-centres on the artificial-feel springs of §2.1. The maintenance source states the job of the device and names the manufactured feel explicitly. Per AMM 27-92-00:

The function of the side sticks is to transmit to the FCPC 1, 2, 3 and to the FCSC 1, 2 the lateral and longitudinal manual control orders in the form of electrical signals, depending on the position of the hand grip, and to generate the related artificial feel loads.

The operational consequence: the force in your hand reflects how far you have moved the stick, not the load the aircraft is carrying. A conventional control loads up with speed and g, giving a natural warning; the sidestick does not — which is exactly why Normal Law has to guard the envelope for you (see 00 §8).

Not mechanically linked / no feedback. The two sticks are independent: you move yours, the other pilot moves his, and neither back-drives the other. The price is that simultaneous inputs do not "fight" mechanically to warn the crew — they are summed electronically and announced by light and an audio "DUAL INPUT". That priority/summing logic is the subject of 16.

Locked at neutral with the autopilot in. With the AP flying, a solenoid locks the stick at neutral — but not solidly. Per AMM 27-92-00:

In autopilot mode, the side sticks are held in the neutral position with a higher load level in order to prevent any unwanted switching to the manual control mode, while keeping the possibility to override the autopilot if required.

The mechanism is the solenoid of §2.1 (12CA1/12CA2): when the AP is engaged the FMGEC energises it, seating two rollers in V-cams on the grip and holding it at neutral. Push past the threshold and the rollers are forced out, the stick frees, and the AP drops. FCOM gives the threshold. Per FCOM DSC-27-20-30:

The autopilot disengages when a pilot applies a force of more than 5 daN in pitch, or 3.5 daN in roll. Note: This is not the recommended method for autopilot disengagement.

So the override works, but it is not the routine way to disconnect (use the sidestick pb). Once you push past the threshold the stick unlocks, the AP disengages, and the AUTO FLT AP OFF alert triggers.

[!warning]- The neutral lock is a "power-to-lock, power-off-to-release" fail-safe — the opposite of what most people assume.

The AMM breaks the lock into five parts: two V-cams on the grip, two rollers, a spring that bears the rollers onto the cams, the solenoid that drives them into the cams when energised, and — the decisive one — a return spring that holds the rollers clear of the cams when the solenoid is de-energised. Per AMM 27-92-00, the assembly includes a return spring which keeps the rollers apart from the cams when the solenoid is de-energized. The implication: if the lock loses power (electrical fault, FMGEC loss) it does not trap your stick at neutral — the return spring frees it and hands you manual authority by default. Many pilots assume a lock is "powered to release"; here it is the reverse, because freezing the sidestick is unacceptable under any failure. The grip wiring reinforces the philosophy — the harness is split into four independent bundles which end up to six separate connectors, so a single break cannot take out the signal path.

The redundancy behind "4 groups of 3". Why so many potentiometers? The design rule is a textbook single-failure-tolerant split. Per AMM 27-92-00:

The side stick design is such that the rupture or disconnection of any one of its parts cannot cause the loss of the totality of the artificial feel loads on one axis or cannot leave free more than one of the four potentiometer groups related to each axis.

In plain terms: break one spring and the axis still has feel — it does not go limp; free one potentiometer group and three remain to send a trustworthy stick position. The same "split into redundant shares" language reappears on the pedals, the speedbrake lever, and the throttle units — it is the unifying grammar of this whole layer.

[!warning]- Misconception: "the stick has a centring spring, so its force tells me what the aircraft is doing."

No. FCOM is explicit — the sticks do not receive feedback from the flight control surfaces. The force is purely spring characteristic and has nothing to do with the current aerodynamic load. You cannot sense the approach to stall or over-g through a stiffening stick — there is none. That sensing is done by Normal Law's protections, not by your hand.

The flying-technique layer — what the FCTM teaches

Everything above is engineering. The FCTM reframes the same facts as operational benefits and a basic skill the crew must actually train. First, the AP-engaged lock seen from the pilot's seat. Per FCTM AOP-10-30-10:

When the autopilot is engaged: ‐ The sidesticks are locked in neutral position (immediate tactile feedback) ‐ There is no possibility of simultaneous input from the flight crew and the autopilot ‐ The autopilot can be disconnected instinctively, at any time, by a firm pressure on the sidestick.

Read what the FCTM does with the mechanical fact: the lock is immediate tactile feedback — rest a hand on the locked, centred, stiffened stick and you feel that the AP is flying, without looking at the FMA. That is the solenoid lock's real value to the pilot, not just "prevent inadvertent reversion". The FCTM then lists the full case for the side-mounted stick. Per FCTM AOP-10-30-10:

The main operational benefits of the side-mounted stick: ‐ It enables a non-obstructed view of the main instrument panel ‐ It is adapted for emergency situations (e.g. incapacitation, stick jamming, control failures) ‐ It fits comfortably into the hand with a correct adjustment of the armrest ‐ It makes the sliding table installation possible (e.g. for maps, documents, meals).

The third benefit — "with a correct adjustment of the armrest" — points at the skill the FCTM stresses most and pilots ignore most: how to set the armrest, wrist and feet. Per FCTM PR-NP-SOP-70:

After adjusting the seat, each pilot should adjust the outboard armrest, so that the forearm rests comfortably on it, when holding the sidestick. There should be no gaps between the pilot's forearm and the armrest. The pilot's wrist should not be bent when holding the sidestick. This ensures that the pilot can accomplish flight maneuvers by moving the wrist instead of lifting the forearm from the armrest. Symptoms of incorrect armrest adjustment include over-controlling, and not being able to make small, precise inputs. The flight crew must have their feet in a position so that full rudder deflection combined with full braking, even differential, can be applied instinctively and without delay.

Tie this back to "no feedback": because the stick gives no aerodynamic cue, the only thing protecting you from over-controlling is your physical posture. A poorly set armrest forces you to lift the whole forearm, you lose the wrist's fine movement, and the FCTM's stated symptoms follow — over-controlling and an inability to make small, precise inputs. On the A330 seating and armrest are a handling-quality matter, not a comfort one. The pedal datum is its own minimum: full rudder plus full differential braking must be reachable instinctively — which leads straight to the pedals.

4. Rudder pedals — rigidly interconnected, but the path to the rudder is electrical

00 set the rule: the A330 rudder is rudder-by-wire. This article adds how the pedal end connects in.

The two pilots' pedals are joined by a rigid mechanical linkage — push the left pedal and the other pilot's left pedal moves with it, as on a conventional aircraft. But that interconnect runs to a pedal transducer, not to the rudder. Per FCOM DSC-27-10-10, Two pairs of rigidly interconnected pedals ensure electrical control of the rudder — the interconnect is between the two pilots' pedals, not a cable to the surface. The transducer turns pedal travel into a signal. Per AMM 27-92-00:

The six identical transducer units transmit: ‐ left and right rudder-pedals position information to the FCPC 1, 2, 3 and the FCSC 1 to compute the lateral laws...

Note the destination: pedal signals reach three PRIM plus SEC 1 — not all five computers. SEC 2 does not receive the pedal signal, unlike the sidestick, which goes to all five. The computers then drive the rudder's three hydraulic servos by the active law; there is no cable anywhere between pedal and rudder. The rudder end of that chain (three parallel servos, travel limit, yaw damper) belongs to 20.

Where the pedal feel comes from. As with the stick, it is manufactured — but the device is a separate box, the PFTU (Pedal Feel and Trim Unit), which handles both pedal feel and rudder trim (trim in §6). Per FCOM DSC-27-10-20:

Inside the PFTU, artificial feel and rudder trim are achieved by two electric motors that position the artificial feel unit.

The transducer hardware (RVDT). The six identical units use a different sensor type from the stick. Per AMM 27-92-00:

The transducer unit consists of a casing with an attachment base, in which are installed: ‐ an input lever whose travel is limited by stops, ‐ two RVDTs driven by the lever, ‐ two electrical connectors, ‐ a built-in rigging device which consists of a hole in the lever in which a pin attached to the casing can be manually engaged.

So note the split across the aircraft: the sidestick uses potentiometers (4×3 per axis), while the pedals, rudder and elevators use RVDTs (two per unit). Two families of position sensor, used deliberately for different jobs — do not merge them into one.

[!warning]- Misconception: "the pedals are connected to the rudder by cables, so it's a hard, direct control."

No — the pedals connect only to a transducer; the rudder is rudder-by-wire. The cable-to-rudder model does not apply. But keep the 00 warning in mind: in Normal Law the rudder has no envelope protection. Even though it is fly-by-wire, large, rapid or reversing pedal inputs are not stopped by the system and can overstress the fin. Fly-by-wire does not mean automatically gentle.

5. The speedbrake lever — one lever, two functions, one cam

The SPEEDBRAKE lever on the pedestal does two jobs with one handle. FCOM gives the cockpit drill. Per FCOM DSC-27-20-30:

SPEEDBRAKE lever Controls: ‐ The position of the speedbrake surfaces. To set speedbrake surfaces to a required position, the speedbrake lever has to be pushed down and set to the required position. A "hardpoint" is provided at "½" SPEEDBRAKE position. ‐ The manual preselection of the ground spoilers. To arm the ground spoilers, the lever must be pulled up when in the RET position.

The two gestures are deliberately orthogonal: push down then aft = extend speedbrake; pull up = arm the ground spoilers. You cannot turn an "arm" into an "extend" by accident. The AMM shows how a single cam delivers both. Per AMM 27-92-00:

The cam profile is such that: . the lever is locked in the "speedbrakes retracted" position under the action of the spring located in the duplicate shaft drive part. A push action on the handle unlocks the lever and permits control of the speedbrake extension by moving the handle in the aft direction. . a hard point is felt at mid stroke of the lever. . a pull action on the lever in the "speedbrakes retracted" position causes rotation of the lever which generates the ground spoiler preselection signal. The colored ring at the base of the lever becomes then visible.

Put together with the lever's states:

• At rest — a spring locks the lever retracted, so you cannot brush the speedbrake out.
• To use speedbrake — push to unlock, then move aft; the hard point at mid-stroke is a tactile detent (about half speedbrake).
• To arm ground spoilers — pull up; the coloured ring at the lever base appears as a visual confirmation that the spoilers are armed.

Redundancy on the signal side. Lever position is read by six middle-point potentiometers and sent to FCPC 1/2/3, with the same split philosophy. Per AMM 27-92-00:

The unit design is such that the rupture or disconnection of any part cannot leave free more than three out of the six potentiometers included in the equipment.

Everything downstream of the lever — how far the surfaces extend, the inhibition conditions, automatic ground-spoiler extension, roll priority over speedbrake — is left to 22 and 23. This article owns only "how the lever turns intent into a signal".

6. The pitch trim wheel — the only mechanical channel, and it never drops a computer

A pitch trim wheel sits on each side of the pedestal. It is the one channel on the aeroplane that does not rely on electrics — the "mechanical tail" 00 keeps naming. FCOM states its two core properties. Per FCOM DSC-27-20-30:

PITCH TRIM wheel Both pitch trim wheels provide mechanical control of the THS and have priority over electrical control. Note: Crew action on the pitch trim wheel does not disconnect the PRIMs (micro-switches, actuated by the override mechanism, ensure that the computers remain synchronized with the manually-selected position).

Two counter-intuitive points hide in there.

1. Mechanical priority over electrical. Turn the wheel and the mechanical run takes the THS, overriding whatever the computers are commanding electrically — even with all five computers healthy. The architecture section gives the availability condition. Per FCOM DSC-27-10-20:

Mechanical control of the THS is available from the pitch trim wheel, at any time, if either the blue or the yellow hydraulic system is functioning. Mechanical control from the pitch trim wheel has priority over electrical control.

"Mechanical control" means the command path from wheel to THS is mechanical — the muscle is still hydraulic (the Blue or Yellow motors). You are not hand-cranking the stabiliser; you are mechanically commanding the hydraulic motors. This is the same control-versus-actuation split as 00: the wheel supplies mechanical control, the actuation stays hydraulic.

2. It does not disconnect the PRIMs. You might expect grabbing trim by hand to drop the computers. It does not. Micro-switches on the override mechanism tell the PRIMs the manually-selected position in real time, so they stay synchronised, raise no fault, and do not degrade. This is why "USE MAN PITCH TRIM" in Direct Law is a clean operation — you trim by hand and the computers follow without a fight.

How the wheel is used. FCOM covers two situations. Per FCOM DSC-27-20-30:

‐ Before takeoff, the pilot sets the THS to the angular value, determined as a function of the aircraft CG, using the CG scale on the wheel. The relationship between the aircraft CG and the THS setting on the trim wheel is only applicable for takeoff. The limits of the THS normal setting range for takeoff are indicated by a green band on the pitch trim wheel. ‐ In flight, when in direct law, the pilot uses the THS conventionally to fly in trim.

Note the easily-misread detail: the wheel's CG scale is valid for takeoff only. In flight, trim depends on CG, weight, altitude and speed, so the "CG value" shown on the wheel no longer corresponds to the real CG — do not read CG off the wheel in the air.

Automatic reset after landing. Per FCOM DSC-27-20-30:

After nosewheel touchdown, when the pitch attitude becomes less than 2.5 ° for more than 5 s, and the ground spoilers are retracted, pitch trim is automatically reset to 4 ° UP.

On configurations that distinguish a touch-and-go, the same logic adds a speed test: below 80 kt the trim resets to the UP value (4° UP, or 5° UP on some configurations), while above 80 kt a touch-and-go is detected and the trim is set back to the takeoff value based on the CG measured during the approach.

[!warning]- The pitch trim wheel is the one input the autopilot never takes away — and it commands hydraulic muscle mechanically.

Two things trip people up. First, "mechanical" does not mean you move the stabiliser by arm-strength — the wheel mechanically commands the Blue/Yellow hydraulic motors; with neither hydraulic system running there is no mechanical trim. Second, manual trim does not disconnect the PRIMs: the override micro-switches keep the computers synchronised, so there is no fault and no law downgrade. Hold this against the sidestick (locked at neutral with AP in) and the rudder trim knob (inactive with AP in): the wheel is the only input that still works, mechanically, with the autopilot engaged.

Configuration note: the post-landing reset value (4° vs 5° UP) and the touch-and-go logic vary by effectivity — defer to your own aircraft configuration and do not pin a single value as "the A330 value".

7. Rudder trim — knob, reset, two motors

The cockpit end of rudder trim is the RUD TRIM knob, a RESET pb and a position indicator on the pedestal. Per FCOM DSC-27-20-30:

RUD TRIM selector Controls the rudder trim actuator, which moves the neutral point of the artificial feel by an equivalent of : In clean configuration : 1 °/s of rudder travel ... RESET pb By pushing the RESET pb, the zero trim position is ordered at 3 °/s. ... RUD TRIM Position Indicator Displays rudder trim direction (L or R) and value (0 to 25 °).

"Trim moves the neutral point of the artificial feel" captures the essence of Airbus trim: rudder trim does not directly deflect the rudder — it shifts the zero-point of the pedal feel, so that with feet off, the rudder settles at the trimmed position. That feel-plus-trim function is the two motors in the PFTU of §4, run by the SEC. Per FCOM DSC-27-10-20:

In normal operation SEC 1/MOTOR 1 operate with SEC 2/MOTOR 2 synchronized as a backup.

A frequently-tested logic: with the autopilot in, the knob does nothing. Per FCOM DSC-27-10-20:

Note: With the autopilot engaged, rudder trim orders are computed by PRIM and transmitted to the SEC for actuation. The rudder trim switch and the rudder trim reset pushbutton are not active.

So with the AP flying, turning the knob has no effect — trim is computed automatically by the PRIM. Put this beside the sidestick (locked at neutral with AP in) and the trim wheel (still mechanically available with AP in) and a pattern emerges: once the AP is engaged, the electrical input devices mostly hand their job to the computers — the mechanical wheel being the lone exception.

Configuration note: the rudder-trim indication range shows as 0 to 25° or up to ±29.2° depending on effectivity — defer to your own configuration. The RUDDER TRIM RUNAWAY procedure is covered in QRH Jam and Loss of Control: hold the wings level laterally, use the rudder with care, centre the rudder with the pedals, and land in a normal configuration.

8. Why pitch trim and rudder trim, but no aileron trim

Lay out the input devices this article has covered: a pitch trim wheel and a RUD TRIM knob on the pedestal — but no roll/aileron trim control of any kind. That absence is a hard fact of the device inventory. Per FCOM DSC-27-10-10:

No manual aileron trim switch is provided.

Why none is needed ties the two stakes from 00 and 10 together — and this next step is reasoning, not a verbatim manual statement:

• In Normal Law the roll axis has automatic bank/spiral trim: release the stick and the aircraft holds the current bank (returning to and holding wings level within the protected band). The system is already trimming roll for you.
• Pitch is different: the aircraft needs a settable mechanical reference (the THS), so the wheel stays. Yaw needs a holdable offload of pedal force in asymmetric/crosswind cases, so rudder trim stays.
• Roll has no standing asymmetric force that needs offloading as a matter of course, and Normal Law already holds bank — so no separate roll-trim control is provided.

The "why none is provided" reasoning above is integrative synthesis from the lateral-law conclusions of 00/10 plus this article's device inventory; the source does not state it in one sentence. The fact to retain is the device-list fact: the A330 cockpit has no roll-trim control — that part is hard and directly citable.

9. The sensor families and the non-pilot inputs

Pull the sensors into focus. The control inputs use two families of position sensor, by job:

• Potentiometers (plastic-track, middle-point) — on the pilot-moved levers: the sidestick (4×3 per axis), the speedbrake lever (six), and the throttle control units. Common signature: duplicate mechanism, middle-point, redundant split.
• RVDTs (two per unit) — on the six identical transducer units: pedals ×2, rudder ×2, elevators ×2.

Beyond the hand- and foot-moved controls, AMM 27-92 groups several inputs that the pilot does not operate but that still feed the computers:

• Vertical/lateral accelerometers and a yaw rate gyro. These are the computers' back-up "senses". Per AMM 27-92-00, the vertical accelerometer data is used in the event of failure of both ADIRUs to consolidate the load factor information supplied by the valid ADIRU and also in turbulence damping; the rate gyro information is used for the elaboration of the Dutch Roll damping law in the event of loss of the 3 ADIRUs. Their role in the laws is deepened in 11 and with the ADIRS (ATA-34).
• Throttle control units ×2. Each sends throttle position to engine control and to the FCPC for ground-spoiler logic, with a binding single-failure rule. Per AMM 27-92-00:

The unit design is such that no single failure is able to cause the loss of drive of the two sets of potentiometers without loss of resolvers.

The point of that rule: the resolvers (engine control) and the potentiometers (flight controls) are tied together in failure — a single fault either leaves both alive or loses both, so you never get the dangerous mismatch of "the engine thinks you closed the throttle while the flight controls think you are still pushing". The ground-spoiler use of this is in 23.

• Hydraulic pressure switches. They tell FCPC1/FCSC1 whether Green/Blue/Yellow are pressurised, with a snap-action disk-spring movement so the switch trips crisply at a set pressure rather than dithering near the threshold. This is the sensor floor under the "Green loss → P2 becomes master" logic of 01.
• PRIM/SEC/TURB DAMP pushbuttons. Guarded engage/reset buttons — pressed in to engage, released to reset to neutral; their law functions are in 01.

10. Power supply — where the electricity comes from

The input devices themselves are mostly passive (springs and sensors); what actually draws power is the computers that read them — they excite the position transducers and read the signal back. So "powering the sidestick" is really "powering the FCPC/FCSC". Per AMM 27-00-00:

In flight or on the ground, with the engines running: The flight controls are supplied with DC power from the essential busbars 4PP, 8PP and the normal busbar 2PP... These bus bars ensure power supply: ‐ to the computers which supply the position transducers and the actuator control components... The FCPC1 and FCSC1 power supply is performed through a PSDU (diode box) (Ref. 27-92-00).

Three things to pull out:

1. The sensor's power comes from the computer. The computers which supply the position transducers — your stick and pedal transducers are not on a busbar directly; they hang off the computers. So one computer losing power affects that computer's reading path, not "the sidestick going dead" — the stick signals all five computers, and the survivors keep reading.
2. FCPC1/FCSC1 go through a PSDU. The diode box decouples normal and battery supplies so these two key computers transition seamlessly onto the battery. Per AMM 27-92-00:

The diode unit permits decoupling of the FCPC1 and FCSC1 normal and battery power supplies. Decoupling is achieved via two power diodes D1 and D2. Diodes P1, P2, P3 are used for lightning strike protection.

3. On the ground on battery, you need hydraulics. Per AMM 27-00-00:

On the ground, with the engines stopped, all the flight controls can be supplied from the APU or from one (or two) electrical ground power unit(s). The FCPC1 and FCSC1 also can be supplied from the batteries if at least one hydraulic system is pressurized. This condition is achieved via relays and a logic built into each computer, which control relay cutoff if the pressures in the three hydraulic systems are detected low.

The logic is clean: battery capacity is precious, and with no hydraulics your electrical control cannot move a surface anyway — so when all three systems read low, the computers cut the battery feed themselves rather than waste it. Busbar reconfiguration (loss of 2PP, double-generator loss) belongs to ATA-24 and 01; the three facts to carry are: transducers are powered by the computers, the two key computers run through a PSDU, and ground battery supply depends on hydraulic pressure.

11. The devices across a flight

Six scenes turn the static inventory into a moving picture:

1. Pre-flight flight-control check — the PF moves the sidestick to the full-travel stops in pitch and roll while the PM checks elevator/aileron deflection and correct spoiler movement on the ECAM F/CTL page (per FCTM PR-NP-SOP-100). This directly proves the stick → transducer → computer → surface chain is intact. On the ground the stick positions are also shown on both PFDs.
2. Takeoff trim set — the wheel is turned to set the THS to the CG value inside the green band; a setting outside the band triggers a takeoff configuration warning, binding the mechanical trim input to takeoff safety.
3. AP engaged in the cruise — the sticks lock at neutral with a raised threshold; the RUD TRIM knob and RESET are inactive (trim computed by PRIM); but the trim wheel can still mechanically override. Three devices, three different "AP-engaged" treatments — exactly the pattern of §7.
4. Engine-out, trim the rudder — full pedal toward the live engine (rigid linkage → transducer → PRIM/SEC → three rudder servos) holds the yaw, then the RUD TRIM knob offloads the pedal force by moving the feel neutral point. Recall the 00 rule: the rudder has no protection — use it with care.
5. Speedbrake on approach — push down to unlock, move aft to the detent at the hard point; before landing, pull up to arm, the coloured ring appears, and the spoilers deploy automatically at touchdown.
6. Rudder trim runaway — the QRH principle (wings level laterally, rudder with care, centre with the pedals, normal landing configuration); the full procedure is in 34.

On the dispatch side, failures of these devices tend to be unforgiving: ECAM messages such as F/CTL L(R) SIDESTICK FAULT, F/CTL PEDAL SENSOR FAULT and F/CTL RUD PEDAL FAULT mostly point to "no dispatch" or to running the ECAM/PRIM1·SEC1 actions per some operators' MEL — reflecting that control inputs are safety-critical and, despite the redundancy, not casually dispatched with a fault. The specific conditions are in the MEL and in EFCS Computer Failures.

Self-test

[!note]- Q1. What do "springloaded", "not mechanically linked" and "no feedback" each mean for the sidestick, and does the stick force reflect the aerodynamic load?

"Springloaded" — release it and it self-centres on the artificial-feel springs. "Not mechanically linked" — the two sticks are uncoupled in both mechanism and electrics, so neither back-drives the other; dual inputs are resolved electronically by priority logic (16). "No feedback" — the stick takes nothing back from the surfaces. So the force does not reflect the aerodynamic load: per FCOM it does not receive feedback from the flight control surfaces, and the force is pure spring characteristic. That is exactly why Normal Law has to guard the envelope — you cannot feel the approach to stall or over-g through a stiffening stick.

[!note]- Q2. Is there a cable between the rudder pedals and the rudder? Which computers receive the pedal signal?

No cable. The two pilots' pedals are rigidly interconnected to each other (synchronised), but that linkage runs only to a pedal transducer — the rudder is rudder-by-wire. The pedal signal reaches FCPC 1/2/3 + FCSC 1; note SEC 2 does not receive it, unlike the sidestick which signals all five. The "cable to the rudder" model does not apply to this aircraft.

[!note]- Q3. How does one speedbrake lever separate "extend speedbrake" from "arm ground spoilers", and what are the hard point and coloured ring?

By gesture plus cam profile: push down to unlock then move aft = extend speedbrake, with a tactile hard point at mid-stroke (about half speedbrake); pull up in the retracted position = arm the ground spoilers, at which point the coloured ring at the lever base appears as a visual confirmation. Two orthogonal gestures prevent confusing one for the other. Lever position is read by six potentiometers feeding FCPC 1/2/3.

[!note]- Q4. The trim wheel is a mechanical channel — why does using it not disconnect the PRIMs, and can it be used with the autopilot in?

The override mechanism drives micro-switches that tell the PRIMs the manually-selected position in real time, so the computers stay synchronised — no fault, no degradation, no disconnect. It has priority over electrical control and is available at any time as long as Blue or Yellow hydraulic is working, including with the AP engaged (where it can still mechanically override). That contrasts with the sidestick — locked at neutral with the AP in — and the RUD TRIM knob — inactive with the AP in: the wheel is the only input the autopilot never takes away.

[!note]- Q5. The A330 has pitch trim and rudder trim but no roll trim — why? And where does the transducers' power come from?

No roll trim because Normal Law already trims roll for you — automatic bank/spiral trim holds the current bank when you release the stick. Pitch needs a settable mechanical reference (THS wheel) and yaw needs a holdable offload of asymmetric pedal force (RUD TRIM); roll has no such standing need, so no control is provided (the absence is a hard device-list fact; the "because" is reasoning). Power: the devices are mostly passive — the position transducers are powered by the computers, which take DC from busbars 4PP/8PP/2PP; FCPC1/FCSC1 come in through a PSDU diode box; on the ground the battery can feed FCPC1/FCSC1 only if at least one hydraulic system is pressurised (all three low → the computers cut the battery feed).

[!note]- Q6. With the autopilot engaged, what happens to the sidestick, the rudder trim knob, and the pitch trim wheel?

The sidestick is locked at neutral with a raised force threshold (immediate tactile feedback; firm pressure of >5 daN pitch / >3.5 daN roll overrides and disconnects the AP — though that is not the recommended method). The RUD TRIM knob and RESET pb are inactive — trim is computed by the PRIM and sent to the SEC. The pitch trim wheel still works mechanically and retains priority over electrical control. The pattern: electrical inputs hand their job to the computers when the AP is in; the mechanical wheel is the exception.

Key takeaways

References

Per FCOM DSC-27-20-30 (Controls and Indicators — sidesticks springloaded/uncoupled/no feedback/AP neutral lock; AP-disengage force 5 daN pitch, 3.5 daN roll; RUD TRIM selector/RESET/indicator; SPEEDBRAKE lever hardpoint and ground-spoiler arming; PITCH TRIM wheel mechanical priority, micro-switches do not disconnect PRIMs, CG green band, post-landing/touch-and-go reset). Per FCOM DSC-27-10-20 (Architecture — THS mechanical control available with Blue or Yellow; PFTU artificial feel and rudder trim by two motors, SEC 1/MOTOR 1 with SEC 2 backup; rudder trim inactive with AP engaged). Per FCOM DSC-27-10-10 (interconnected pedals ensure electrical control of the rudder; no manual aileron trim switch). Per AMM 27-92-00 (sidestick function and artificial feel; AP-mode hold with override; single-failure split of feel loads and potentiometer groups; V-cam/return-spring fail-safe lock and four-bundle/six-connector harness; six identical RVDT transducer units to FCPC 1/2/3 + FCSC 1; speedbrake cam profile, coloured ring and six-potentiometer split; throttle resolver/potentiometer single-failure rule; accelerometer and rate-gyro back-up of load factor and Dutch-roll damping; PSDU decoupling). Per AMM 27-00-00 (electrical power supply — DC busbars 4PP/8PP/2PP, computers supply the transducers, FCPC1/FCSC1 via PSDU, ground battery supply conditional on at least one hydraulic system pressurised). Per FCTM AOP-10-30-10 (operational benefits of the side-mounted stick; AP-engaged tactile feedback and instinctive disconnect) and FCTM PR-NP-SOP-70 (seating, armrest and pedal adjustment technique). Per FCTM PR-NP-SOP-100 (pre-flight full-travel flight-control check). Rudder-trim runaway handling per the QRH (Rudder Trim Runaway), full procedure in 34. The "why no aileron trim" rationale and the "AP hands off to the computers" pattern are integrative synthesis built on the cited facts, not single verbatim manual statements; configuration-dependent values (post-landing reset 4°/5° UP, rudder-trim range 25°/29.2°) are flagged for deferral to the applicable aircraft configuration.

Independent study material, not an Airbus publication and not endorsed by the manufacturer. Always defer to the current operator FCOM, FCTM, and QRH for operational use.