Air Cycle Machine — Engineering Deep Dive

Pack Principles gave the air cycle as a concept — compress, reject heat, expand, with the turbine driving the compressor and fan on one shaft. This deep-dive opens up the machine itself: its coaxial layout, bearings, blade materials, the four heat exchangers and the water loop inside the pack, how it cools itself without oil, why the turbine outlet must be protected against icing, and what is left when it fails. The pack controller's own internals (channel election, BITE, the ARINC bus) are in Pack Controller; the PFCV was ata-21-06.

1. Location

The air conditioning packs 521HH (522HH) are the same and are installed in the unpressurized area of the belly fairing behind FR 40. — AMM 21-52-00 §3.A (English fact source; the Chinese FCOM carries no AMM content)

The packs (FIN 521HH / 522HH) are identical, in the unpressurised belly fairing behind frame 40. Two consequences: the pack's high-temperature path (compressor outlet ~200 °C class) is outside the pressure vessel, so a pack hot-air leak vents to the belly, not the cabin; and pack 1 is on the left (zone 191), pack 2 on the right (zone 192), physically separated to prevent common-cause failure. Maintenance enters through access panels 191/192.

2. The internal airflow — full single-line (FIN-annotated)

        bleed (≈200 °C, ≈30–40 psi)
              │
              ▼
        [PFCV 511HB/512HB]   ── covered in ata-21-06
              │
              ▼
   ┌──[Primary HX 521HH18]───┐ ◄── ram air
   │  PHX, plate-fin          │
   │  crossflow, aluminium    │  ↓ ΔT ≈ −100 °C
   └──────────┬───────────────┘
              ▼
   ┌──[Compressor 521HH12]────┐
   │  Ti impeller, 17 blades   │  ↑ pressure + temperature
   └──────────┬───────────────┘  ↑ to ~200 °C class
              ▼
   ┌──[Main HX 521HH19]───────┐ ◄── ram air
   │  MHX, plate-fin           │
   │  crossflow, aluminium     │  ↓ ΔT ≈ −50…80 °C
   └──────────┬───────────────┘
              ▼
   ┌──[Reheater 521HH15 hot]──┐
   │  plate-fin crossflow      │
   └──────────┬───────────────┘
              ▼
   ┌──[Condenser 521HH13 hot]─┐
   │  tube crossflow; vapour   │
   │  condenses to droplets    │
   └──────────┬───────────────┘
              │ water-laden
              ▼
   ┌──[Water extractor 521HH14]┐
   │  swirl vanes → wall →      │──► water injector 5531HB
   │  sump → drain              │    (sprays ram air for cooling)
   └──────────┬────────────────┘
              │ dry cold air
              ▼
   ┌──[Reheater 521HH15 cold]─┐
   │  reheats to prevent       │
   │  downstream icing         │
   └──────────┬───────────────┘
              ▼
   ┌──[Turbine 521HH12]───────┐
   │  Al cast, 10 full + 10    │
   │  splitter blades          │
   │  ↓ expands → drives shaft │
   │  ↓ to ~0 °C or below      │
   │  ◄── anti-ice inlet (hot) │
   └──────────┬───────────────┘
              ▼
   ┌──[Condenser 521HH13 cold]┐
   │  cold air absorbs heat    │
   │  from the hot side →      │
   │  warms toward outlet      │
   └──────────┬───────────────┘
              │ ◄── temperature control valve 521HH3 (bypass hot air)
              ▼
   ┌──[Downstream check valve 5533HB]┐
   │  one-way                         │
   └──────────┬───────────────────────┘
              ▼
        [mixing unit] → cabin

Sources: AMM 21-52-00 §3.A (general + component-location table) + FCOM DSC-21-10-20 (pack schematic).

Two counter-intuitive points in this layout:

• The water loop passes the reheater twice. The same reheater warms the air before water extraction (hot side) and after (cold side) — using main-airflow waste heat for two jobs, saving a heat source.
• The condenser carries two streams at once. The reheater-out air (still water-laden) goes through its hot side; the very cold turbine outlet goes through its cold side. The cold air cools the hot air → vapour condenses → then the cold air absorbs that heat and warms toward the pack outlet. The condenser both cools and re-warms — the most ingenious part inside the pack.

3. The coaxial three-wheel topology

   ┌────────────────────────────────────────┐
   │   Air Cycle Machine 521HH12 (522HH12)  │
   │                                        │
   pitot ──► cooling air → journal bearings  │
   (ext.)    + maintains positive pressure   │
   │                                        │
   │   2 × 50.8 mm (2.0000 in) journal bearings
   │                  │                      │
   │                  ▼                      │
   │   ┌──────────────────────────────────┐ │
   │   │           one shaft              │ │
   │   │  [compressor wheel] Ti forged,   │ │  ← driven by the turbine
   │   │       17 radial blades           │ │
   │   │  [turbine wheel] Al cast,        │ │  ← the ONLY power source
   │   │       10 full + 10 splitter      │ │
   │   │  [fan wheel] Ti forged,          │ │  ← driven by the turbine,
   │   │       17 radial blades           │ │     drives ram air
   │   └──────────────────────────────────┘ │
   │   thrust bearing absorbs axial loads    │
   │   housing: aluminium alloy + stainless  │
   └────────────────────────────────────────┘

The air cycle machines 521HH12 (522HH12) have a compressor wheel, a turbine wheel and a fan wheel. All of the wheels are installed on the same shaft. The shaft is installed in 50.8 mm (2.0000 in.) journal bearings. A thrust bearing absorbs axial loads. Air supply from a pitot tube decreases the temperature of the journal bearings. It also makes sure that the journal bearings have a positive pressure. There is an anti-ice inlet to add hot air to prevent ice in the turbine outlet. — AMM 21-52-00 §5.A

The turbine wheel is an aluminum investment casting. It has ten full and ten splitter blades. The compressor wheel and the fan wheel are machined titanium forgings. They are radial flow impellers and have 17 blades. The housing of the air cycle machine is made of aluminum alloy and corrosion-resistant steel. — AMM 21-52-00 §5.A

[!warning]- Coaxial ≠ all "driven"

Do not read "compressor" as the engine-driven kind. Inside the ACM the turbine is the only power source — high-pressure bleed expands through it and the torque drives the one shaft. On that shaft: the compressor wheel (driven by the turbine, giving a second-stage compression to the primary-HX-cooled air — a load, not a source) and the fan wheel (driven by the turbine, forcing ram air through the heat exchangers — a second load). Energy account: bleed heat → turbine work → fed back to the compressor + fan. The ACM draws no extra engine mechanical power (there is no mechanical interface after the PFCV); it self-balances: turbine work = compressor draw + fan draw + friction.

[!warning]- The journal bearings are cooled by external pitot air — there is no oil

The ACM has no oil circulation (many assume it has an oil pump like an engine). The bearings are an air-bearing design: a pitot-tube air feed enters the bearing cavity to (1) carry away friction heat and (2) hold the cavity at positive pressure so internal high-temperature air is kept out of the bearing cavity (otherwise the hot turbine-end air would destroy the bearings). On the ground (no ram effect) the bearing cooling comes via the ram-air system's equivalent path; after the takeoff roll the pitot is effective. The ACM cannot run at high load for long with a blocked/failed pitot — it fails from the bearing end first. This is why a pitot-probe heat failure is a real problem — not just an airspeed-indication issue.

[!warning]- The turbine outlet is "heated to prevent ice" — what is added is not a refrigerant

Turbine-outlet temperature can fall to −30…−50 °C class (exact value set by bleed inlet temperature, ACM speed, and HX efficiency — the FCOM gives no formula) — below the freezing point of water. If upstream water is not fully removed, it freezes at the turbine outlet, blocks the flow path, pack flow drops, snowballing. The anti-ice inlet injects hot air (tapped after the PFCV, before the primary HX) — a small uncooled stream mixed into the turbine outlet to lift it above freezing. This is not "turning on refrigeration" — the A330 pack uses no refrigerant; this path fights ice with heat. The crew sees no indication for it normally; a failure shows as abnormal pack-outlet temperature + downstream water-extractor blockage, found at maintenance as ice traces at the turbine outlet. AMM task 21-52-00-960-802-A specifically discards the flexible hose between compressor and turbine before its aging delaminates and blocks the water extractor — hence its scheduled discard in the MPD.

4. The working cycle

The two packs operate automatically and independently of each other. ... Warm pre-conditioned bleed air enters the cooling path via the pack flow control valve and is ducted to the primary heat exchanger. Then, the cooled bleed air enters the compressor section ... and is compressed to a higher pressure and temperature. It is again cooled in the main heat exchanger, and enters the turbine section where it expands. In expanding, it generates power to drive the compressor and cooling air fan. The removal of energy during this process reduces the air temperature, resulting in a very low air temperature at turbine discharge. The temperature control valve can modify the pack outlet temperature by adding uncooled air to the turbine outlet flow. In case of an air cycle machine failure, bleed air is only cooled by the corresponding heat exchanger only. — FCOM DSC-21-10-20

One-line summary: the ACM is a heat source + refrigerator + drive in one — no external drive (no mechanical interface after the PFCV), no refrigerant. The turbine converts the bleed's enthalpy into shaft work; the shaft work feeds the compressor (second-stage compression) and the fan (ram air through the heat exchangers). The whole budget: cooling capacity = bleed inlet enthalpy − turbine outlet enthalpy. After an ACM failure only the heat exchangers remain → cooling is just "how much the ram air pulls the bleed down".

[!note]- The fan is active only on the ground

The expansion of the air in the turbine turns the turbine wheel, the compressor wheel and the fan wheel. The fan wheel gives a flow of ram air through the ram air system if there is no ram air effect (on the ground). — AMM 21-52-00 §3.C

Pilots tend to think "the fan is always blowing". In flight the natural ram flow exceeds what the fan can drive — the fan is then turned by the ram air (a passive role); only on the ground (and slow taxi) does the fan actively blow. A check valve in the plenum (§8) bypasses the fan in flight when the ram flow is large, so the fan does not become an obstruction.

5. Compressor and turbine sections

Air enters the compressor from the primary heat exchanger and is compressed. The pressure and temperature increase. The air then flows to the main heat exchanger. — AMM 21-52-00 §3.C

Why a second compression: bleed arrives ~200 °C class at tens of psi; the primary HX cools it to ~100 °C class (rejecting heat to ram air); the compressor compresses it again → temperature back up to ~200 °C but at higher pressure; the main HX then cools it — and high-pressure air rejects heat more efficiently in a heat exchanger (larger temperature-difference driving force, higher transfer coefficient). After the main HX the air can be lower in temperature than the primary-HX inlet but higher in pressure — and that pressure is what lets the turbine extract more work (work scales with the pressure ratio). The titanium impeller is chosen because at ~200 °C aluminium is near softening; titanium gives heat resistance + strength at lower weight. 17 radial blades = a high single-stage pressure ratio in a compact form.

Air enters the turbine section of the air cycle machine from the reheater and is expanded. The pressure and temperature decrease. The air then flows to the condenser. — AMM 21-52-00 §3.C

The turbine is aluminium cast because its inlet temperature is lower than the compressor's (already cooled by the main HX + water loop, ~30–50 °C class) — aluminium suffices and is far cheaper than titanium (turbine blades are a consumable). 10 full + 10 splitter blades is the classic turbine layout: the 10 full blades set the overall aerodynamics; the 10 short splitter blades sit between them in the rear of the wheel to add blade area and suppress flow separation, raising efficiency — total blade area near 20 but the inlet throat area = 10 (good for high-pressure admission at the inlet, distributed expansion in the rear).

6. Heat exchangers — four plate-fin crossflow units

The primary heat exchangers 521HH18 (522HH18) are made of aluminum alloy. They are plate and fin type and have a single-pass crossflow configuration. ... The main heat exchangers 521HH19 (522HH19) are made of aluminum alloy. They are plate and fin type and have a single-pass crossflow configuration. ... — AMM 21-52-00 §5.F

Plate-and-fin = two fluids separated by plates, each fluid side finned to enlarge the transfer area; crossflow = the two flows are perpendicular (between counterflow and parallel in efficiency, but compact — suiting the belly-fairing space); single-pass = straight through, no return bend.

The Ram air flows through the heat exchangers and decreases the temperature of the hot bleed air from the pneumatic system. (primary HX) Ram air flows through and decreases the temperature of the hot air from the compressor of the air cycle machine. (main HX) — AMM 21-52-00 §3.B + §3.D

The two heat exchangers share one ram-air system: ram air enters, passes the main HX first (the segment nearest the turbine inlet — the critical one) then the primary HX (the less critical one), and exits. Note the FCOM describes "primary then main" (bleed's perspective); the AMM describes "main then primary" (ram air's perspective) — the two flows run opposite, giving the crossflow some of the counterflow advantage, and putting the cooler ram air on the more critical main HX.

7. The water loop — reheater + condenser + water extractor

The reheaters 521HH15 (522HH15) are installed between the main heat exchangers and the condensers. The hot air from the main heat exchangers increases the temperature of the cold air from the water extractors. The condensers 521HH13 (522HH13) ... The cold air from the turbine ... decreases the temperature of the hot air from the reheater. The temperature of the hot air decreases to less than its dew point and the water in the air condenses. Water extractor 521HH14 (522HH14) is installed between condenser 521HH13 and reheater 521HH15. It removes the water that condenses in the condenser. The condensed water ... drain to water injector 5531HB (5532HB). — AMM 21-52-00 §3.E / §3.F / §3.G

The water extractors ... have a body, an inlet and an outlet. There is a port for a pack temperature sensor 521HH8 (522HH8) on the outlet. ... There are swirl vanes inside that push the water drops to the inner surface and then to the sump and drain port. — AMM 21-52-00 §5.C

Three judgements the pilot needs:

1. Why extract water before the turbine — the turbine outlet is cold; any water there freezes instantly and blocks the path. So extraction first is mandatory.
2. Why reheat the dry air before the turbine — pure cold air into the turbine would make the inlet too cold; expansion produces an even colder outlet → higher icing risk. The reheater uses post-main-HX waste heat to warm the dry air, trading a little cooling capacity for a non-icing turbine outlet.
3. Where the water goes — the water injector sprays it as mist into the ram air (the heat exchangers' external cooling air); the water evaporates, absorbing heat → ram air cooler → heat-exchanger efficiency up. The pack's waste water becomes a cooling resource, especially useful on the ground or a hot takeoff.

[!note]- Counter-intuitive: the pack temperature sensor is at the water-extractor outlet, not at the pack outlet

Intuition says the pack-outlet temperature is measured "at the pack outlet". In fact the AMM places the pack temperature sensor 521HH8 at the water-extractor outlet (after water extraction, before the reheater cold side). Why: this point is the true cold-section outlet temperature, not yet disturbed by the temperature-control-valve bypass heat — the clean feedback for the pack controller. The real pack outlet = what 521HH8 reads + the heat added by the temperature control valve. So when the ECAM pack-outlet temperature reads abnormally high, it is either ① a cold-section failure (HX blocked / ACM failed) or ② the temperature control valve stuck too far open — distinguished by whether bypass appears and by the compressor-outlet temperature trend.

8. The plenum — the ram-air buffer to the ACM

The plenum 521HH17 (522HH17) is installed in the ram air system and is connected to the ACM ... On ground, the fan of the ACM causes air to flow through the plenum. In flight, the flow of ram air is more than the quantity of air which can flow through the fan. Thus the plenum has a check valve to let the ram air bypass the fan. The plenums ... are made of fiber reinforced plastic and have ... a diffusor and a check valve for bypass flow. The diffusor has an acoustic treatment on it to decrease noise. — AMM 21-52-00 §3.J + §5.E

The plenum (FIN 521HH17 / 522HH17, fibre-reinforced plastic) is the ram-air system's buffer to the ACM. Its check valve bypasses the fan in flight (when ram flow is large), so the fan does not become drag; on the ground the fan drives flow through it. Its diffusor has acoustic treatment to cut the high-speed fan noise transmitted into the airframe (part of the cabin noise floor). FRP is chosen because the ram air at the plenum has already been warmed by the heat exchangers — aluminium would be heavier and costlier than needed.

9. The anti-ice inlet

[!note]+ Anti-ice inlet working principle (integrative reasoning, not a verbatim quote)

The AMM states the ACM has an anti-ice inlet but gives no detailed control law. Synthesising AMM 21-52 + 21-53 + FCOM DSC-21-10-40: the air is tapped after the PFCV, before the primary HX (still ~200 °C class); it is injected into the diffuser/nozzle ahead of the turbine inlet; it is modulated by the pack controller on the turbine-outlet temperature signal (the valve is on the anti-ice valve in AMM 21-53); it opens when the turbine outlet runs too cold (near freezing) and closes/cracks in normal cruise (a little flow is enough). It is not a separate subsystem — it is the low-temperature-protection branch of the temperature control function, the lowest-level fail-safe in the pack controller's automatic logic.

[!warning]- Anti-ice failure degrades slowly, not with an immediate ECAM

If the anti-ice valve sticks closed, or the pack controller loses the turbine-outlet temperature reading → the turbine outlet ices → the water extractor upstream blocks → pack outlet flow drops → ECAM shows PACK FAULT or abnormal pack flow. What the crew sees is pack flow hunting / unstable pack-outlet temperature, not necessarily an overt overheat warning. Maintenance finds it as ice traces at the turbine outlet or a blocked water extractor — triggering a hose + anti-ice-valve + pack-controller investigation. MPD 21-52-00-960-802-A discards the compressor-to-turbine flexible hose periodically, precisely to replace it before aging delaminates it and blocks the water extractor.

10. Temperature control valve — the last trim of the pack outlet

The temperature control valve can modify the pack outlet temperature by adding uncooled air to the turbine outlet flow. — FCOM DSC-21-10-20 (FIN 521HH3 / 522HH3 per the AMM 21-53 system)

The pack controller's loop: takes the cabin demand from the zone controller; reads the current pack temperature sensor 521HH8 (water-extractor outlet); computes the gap to target; commands the temperature control valve open/close (big gap → more bypass heat, small gap → less); in parallel modulates the ram-air inlet/outlet flaps (ram flow → HX cooling strength). The two together hold the pack outlet in the comfort band.

[!warning]- Counter-intuitive: a 24 °C cabin target does NOT mean a 24 °C pack outlet

Cabin target 24 °C → the pack outlet is not 24 °C. The pack delivers ≈ 10–20 °C (depending on cabin-temperature feedback), because the pack is a cold source (it must deliver air colder than the cabin target); after the pack, recirculated air (≈ cabin temperature) mixes in at the mixing unit → mixing-unit output ≈ cabin target ± a few degrees; then the hot-air / trim-air system trims per zone (ata-21-09). So a pack outlet above ~25 °C is itself abnormal — it means the cold section has failed.

11. Downstream check valve

The downstream check valves 5533HB (5534HB) ... have a housing and a flap. The flap can open in one direction only. A spring holds the flap in the closed position if the upstream pressure is less than the downstream pressure. — AMM 21-52-00 §5.G

Two roles: when a pack is off/failed, the already-mixed mixing-unit air does not back-flow into the pack (no contamination, condensation, or reverse-driving of the turbine); and when pack 1 fails while pack 2 runs, pack 2's air does not back-feed into pack 1 (no hot/cold cross-flow).

12. ACM failure — bypass mode

If the ACM fails (compressor/turbine seizure), the affected pack may be operated in the heat-exchanger cooling mode. ... the compressor check valve opens, and air is cooled only by the heat exchanger. The ACM seizure reduces the pack flow. — FCOM DSC-21-10-40

If an air cycle machine 521HH12 (522HH12) is unserviceable, the air flows through the primary heat exchanger ..., the pack check valve 521HH21 ... and the main heat exchanger .... Then the air flows through the water extraction loop and the stopped turbine wheel to the pack outlet and further to the mixer unit. A part of the airflow also bypasses the ACM through the temperature control valve 521HH3 ... to the turbine outlet. This is called pure heat exchanger mode because only the heat exchangers decrease the temperature of the air. Only the split duct 521HH20 ... removes water from the air flow. The pressure in the pack decreases. — AMM 21-52-00 §6.B (1)

Bypass mode is not a total pack loss — it still delivers air, slightly cooled, into the cabin. But with reduced cooling, pressure, and flow, a single pack in bypass can hold one cabin zone for a while; both packs in bypass should be treated as a dual-pack failure (cooling is minimal even without an explicit dual-fault ECAM). ECAM: AIR PACK 1(2) REGUL FAULT + PACK 1(2) IN BYPASS MODE.

13. Pack controller channels

Each Controller has a Channel 1 and a Channel 2. One channel is active and the other is in standby. After each touchdown, the active channel changes. Each Controller has a Channel 1 (that is normally in control), and a Channel 2 (that acts as a backup, if Channel 1 fails). (alternate effectivity) — FCOM DSC-21-10-40

Two effectivities coexist: older — channel 1 is always primary, channel 2 takes over only on channel-1 failure; newer (most) — the active channel swaps every landing so both channels are used regularly, preventing the standby from aging unused (the same philosophy as the CPC's landing swap, ata-21-11). Check your variant; for day-to-day understanding, assume the landing-swap behaviour.

PACK CONTROLLERS. CHANNEL 1 OR 2 FAILURE. A Channel 1 or 2 failure has no effect on pack regulation. CHANNELS 1 AND 2 FAILURE. The corresponding anti-ice valve regulates the pack outlet temperature between approximately 9 °C and 15 °C ... The ECAM signals, associated with the corresponding pack, are lost. The flow control valve pneumatically regulates the pack flow to approximately 120 % of the NORM flow. — FCOM DSC-21-10-40

[!warning]- Counter-intuitive: a dual-channel failure gives MORE flow (120 %)

A dual pack-controller-channel failure raises flow to 120 % NORM — because the PFCV's power-loss default is to open at high flow; losing both electrical channels = the PFCV reverts to "pneumatic + power-loss default" = ~120 %. The pneumatic regulation is a low-precision control referenced to cabin pressure. The pack outlet locks to 9–15 °C because the anti-ice valve backstops it (prevents turbine-outlet ice but does not deliver over-cold air). Cabin feel: cooler but high flow — the crew can offset by adding hot/trim air via the zone controller.

[!warning]- A single channel failure gives NO ECAM warning

The FCOM says a single-channel failure has "no effect" — so the crew cannot know it from ECAM; it surfaces only at a maintenance BITE test. A long-standing single-channel state may therefore exist unknown to the crew; only a dual-channel failure is annunciated.

14. Zone controller dual failure

ZONE CONTROLLER. CHANNEL 1 OR 2 FAILURE. ... has no effect ... CHANNELS 1 AND 2 FAILURE. Optimized and backup temperature regulation are lost. The packs deliver a fixed pack outlet temperature of 20 °C (68 °F). A Channel 1 and 2 failure removes all information from the COND SD page, which then displays "PACK REG". Flow selection from the PACK FLOW selector is lost. — FCOM DSC-21-10-40

In practice: a zone-controller dual failure loses all zone temperature control — the crew can only tell passengers "it is 20 °C today"; a pack-controller dual failure can still be trimmed via hot/trim air (if the zone controller is alive).

15. Overheat thresholds — from the ACM's view

The three compressor-outlet thresholds (detailed in ata-21-06 §4.1, with the chain in ata-21-08):

Pack-outlet overheat: the pack temperature sensor 521HH8 (a thermistor at the water-extractor outlet); ECAM amber > 95 °C / clears < 60 °C. To the crew this is AIR PACK 1(2) OVHT. Note the two overheats cannot be told apart from ECAM alone — read the pack-flow trend, whether bypass or anti-ice is co-occurring, and which field (compressor-outlet vs pack-outlet) is amber.

16. Hose aging → water-extractor block → turbine ice (the MPD lesson)

This task is necessary to prevent flexible hose delamination which can cause a blockage in the water extractor system. A blockage in the water extractor system can cause ice building on the ACM turbine. — AMM 21-52-00-960-802-A (Discard of Flexible Hose)

What a pilot takes from this MPD task: there is a flexible hose between the compressor and the turbine (a detail the FCOM does not draw); it ages → the inner wall delaminates → debris blocks the water extractor downstream → condensed water cannot drain → moisture reaches the turbine → the turbine outlet ices. The icing is gradual — not a single event, but an accumulation over cruise as water is not fully removed. If the MPD task is overdue, pack flow slowly drops / pack-outlet temperature drifts late in a cruise. Crew action: a downward pack-flow trend or recurring pack-outlet-temperature wander → write it up so maintenance checks the MPD 21-52-00 hose status.

17. Three-state diagnosis

Diagnosis technique: flow first — 0 = PFCV closed / pack off; locked 120 % = pack-controller dual failure; falling = ACM failed. Pack-outlet temperature next — abnormally high = ACM failed (heat-exchanger-only cooling is weak); locked 9–15 °C = pack-controller dual failure. ECAM completeness last — all pack-1 signals lost = pack-controller dual failure; partial ECAM = ACM failed / PFCV fault.

Self-test

[!note]- Q1. Are the compressor, turbine, and fan independently driven?

No. All three are on one shaft, and the turbine is the only power source (high-pressure bleed expanding). It drives the compressor (second-stage compression) and the fan (ram air, on the ground). The compressor and fan are loads. Energy account: bleed enthalpy → turbine shaft work → compressor draw + fan draw + friction.

[!note]- Q2. How does the ACM cool itself — does it need oil?

No oil. It is an air-bearing design — a pitot-tube air feed (via an equivalent path on the ground) enters the bearing cavity to cool the journal bearings and hold a positive cavity pressure (keeping the hot turbine-end air out). This is why a pitot-probe heat failure is a real issue, not just an airspeed-indication one.

[!note]- Q3. Why does the turbine outlet ice, and how is it prevented?

Why: turbine expansion drops the outlet to −30…−50 °C class, far below freezing. Prevention: the anti-ice inlet taps uncooled hot air (after the PFCV, before the primary HX) into the turbine inlet to lift the outlet above freezing, modulated by the pack controller. Upstream, the water loop (reheater + condenser + water extractor) removes water before the turbine to minimise what reaches it.

[!note]- Q4. Can the pack still be used after an ACM failure (bypass mode), and what ECAM does the crew see?

Yes — air flows through the primary HX → pack check valve → main HX → water loop → the stopped turbine wheel → pack outlet → mixing unit; part bypasses the ACM via the temperature control valve. Only the heat exchangers cool (pure heat-exchanger mode); pack flow drops, pack-outlet temperature rises. ECAM: AIR PACK 1(2) REGUL FAULT + PACK 1(2) IN BYPASS MODE. A single pack in bypass can continue; both in bypass should be handled as a dual-pack failure.

[!note]- Q5. With both pack-controller channels failed, can the pack still control temperature, and what is the flow?

The pack still delivers air — the anti-ice valve takes over and locks the outlet at 9–15 °C; the PFCV reverts to pneumatic, locked ~120 % NORM. The pack's ECAM signals are lost. Distinguishing tell: flow goes up to 120 % (vs ACM failure where flow drops), and the outlet locks cold (vs ACM failure where it rises).

Key takeaways

Common misconceptions

Scope — what this deep-dive covers and defers

References

A330 specifics per FCOM DSC-21-10-20 (the pack working cycle and the ACM-failure heat-exchanger mode), DSC-21-10-40 (pack-controller and zone-controller channel logic and dual-failure behaviour, the ACM-failure mode), and DSC-21-10-50 (the overheat thresholds — cross-referenced from ata-21-06). The ACM physical detail (coaxial wheels, 50.8 mm journal bearings, blade counts and materials, the four plate-fin/tube heat exchangers, the reheater/condenser/water-extractor loop, the plenum, the downstream check valve, the bypass-mode flow path, and the flexible-hose MPD discard) per AMM 21-52-00 §3/§5/§6 and AMM 21-52-00-960-802-A — the English AMM being the fact source where the Chinese FCOM carries no AMM content; AMM quotes are verbatim English with translation for the Chinese edition only. The energy account, the anti-ice control-law synthesis, the bypass-mode performance figures, and the diagnosis technique are integrative syntheses. All engineering detail is from the A330 knowledge base; no cross-type comparison is made.

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.