Abnormal II: Regulation and Monitoring Faults

The leak family (article 08) lives by isolate and never revive. This family lives by the opposite creed: most of these can be rescued, and each rescue has its own technique — a reset, a demand cut, a throttle push, or a human hand replacing a dead computer. The chapter's first principle, stated once and used eight times: parameter sickness can be reset; physical sickness cannot. A pressure or temperature that wandered over a limit may wander back; melted eutectic salt reports a hot spot that no button-cycling will cool.

The spectrum:

1. ENG BLEED FAULT — the one legitimate reset in the chapter

Trigger, per FCOM PRO-ABN-AIR:

This alert triggers when: ‐ The bleed air overheats, or ‐ The bleed air pressure is too low or too high.

All three diagnoses were quantified in article 05: overheat on the 257/270/290 °C ladder, overpressure at >60 psig (closure at 15 s, alert confirmation at 25 s), under-regulation per the three-row PRV LOW REGULATION table. By alert time the BMC has already closed the valve.

The procedure: ENG BLEED (AFFECTED) — OFF THEN ON; if unsuccessful, OFF. And its two footnotes carry the physics:

The FAULT light extinguishes when the failure disappears (overheat or overpressure). PACK FLOW is limited to 80 %.

Why a reset is even legal here: the panel logic from article 05 — selecting OFF resets the FAULT light and the autoclosure signal. Overpressure and overheat are frequently transients (a demand spike, a moment of regulation drift); the cause passes but the latched closure keeps the valve pinned. OFF wipes the latch; ON lets the valve try again with a clean record. Succeed, and the 80 % pack-flow limit is the convalescence diet — less flow through a recently feverish precooler. Fail, and the side goes dark honestly: OFF, and onward to the ABNORM BLEED CONFIG decision (open or closed per the contagion test — section 5 here versus article 08).

The sub-legend that changes everything. Per FCOM PRO-ABN-AIR:

This alert triggers in the case of engine 1(2) bleed overtemperature with the associated engine bleed valve failed open.

Overtemperature plus a valve that refused its electrical closure: hot air is now un-shut-off-able, which morally reclassifies the event as a leak. The procedure borrows the leak playbook wholesale — ENG BLEED OFF (no THEN ON — the valve ignored one closure already), engine-1 case adds APU BLEED OFF, referral to X BLEED Closed (this is the fifth of article 08's five CLOSE conditions), and downstream the users get bricked up (pack off, anti-ice off) inside the CLOSED wrap-up. Inoperative systems: the affected bleed, plus APU BLEED on the number-1 side.

2. HI TEMP — the only temperature alert that touches no valve

Trigger, per FCOM PRO-ABN-AIR:

This alert triggers when the precooler outlet temperature is above 240 °C and when: ‐ The wing anti-ice is activated above FL 320, or ‐ The PACK FLOW selector is not in the LO position, or ‐ The cargo COOLING selector is in the MAX position.

Place 240 °C on article 03's ladder: above 235 (the solenoid's pressure cut-back has already begun) but below 257 (the closure countdown has not). The alert lives in the buffer zone, and its second line is a list of three ways of demanding too much — anti-ice in thin air, full pack flow, cargo cooling at maximum. So the treatment is pure demand management:

Both bleeds operative: PACK FLOW — LO; CARGO COOLING — NORM; if the warning persists, WING ANTI ICE — OFF with AVOID ICING CONDITIONS. One bleed already inoperative: the opposite pack — OFF, wing anti-ice off, avoid icing.

[!warning]- Note what the procedure never says Not one line touches an ENG BLEED pushbutton. A feverish bleed is not executed; its customers are rationed — flow drops, the precooler breathes, the temperature walks itself back down. The "opposite pack" in the single-bleed case is the elegant touch: the surviving bleed is carrying the whole aircraft through the crossbleed, so removing the pack it feeds directly is the biggest single relief available. Consistently, some operators' MELs have no dispatch item for HI TEMP at all — perform the procedure and go; nothing is broken (article 11).

3. LO TEMP — the chapter's only throttle-push

Trigger, per FCOM PRO-ABN-AIR:

This alert triggers when the engine bleed supplies bleed air at a temperature below 150 °C in flight and the WING A-ICE pb-sw is set to ON.

Cool air is only a disease when something needs it hot — the anti-ice condition is built into the trigger, and the sub-legend ON BLEED 1(2) names the cold side. The remedy is unique in this chapter — ENG PWR INCREASE, with its technique note:

The thrust lever of the affected engine must be advanced with the autothrust OFF. Low bleed temperature may be due to low outside air temperature. Therefore, increasing engine thrust may increase bleed temperature and clear the ECAM caution.

Why autothrust must go: A/THR is a symmetric servant — push one lever and it pulls both back to the couple's target. This prescription is asymmetric — more RPM on the cold engine only. Why it works: offtake temperature rides compressor speed (article 02) — spool up and the source itself runs hotter, often enough to clear the alert entirely, because the root cause is usually nothing more sinister than idle thrust on a brutally cold day (article 03's winter note is the same physics on the ground).

Branches: one side affected and the push fails → that side's ENG BLEED OFF (and the crossbleed question resolves OPEN — cold is not on the contagion list). Both sides low → WING ANTI ICE OFF, AVOID ICING, and under severe accretion the familiar VLS +10 / G DOT floor — two cold sides usually mean an environment problem two throttles cannot outrun.

4. Three quiet ones: NOT CLSD, HPV NOT OPEN, BMC FAULT

ENG BLEED NOT CLSD. Trigger, per FCOM PRO-ABN-AIR:

This alert triggers when the engine bleed valve is not automatically closed during the engine start or when the APU BLEED pb-sw is set to ON.

Two entries from article 02's closure inventory failed to execute. The action is the human fallback — ENG BLEED (AFFECTED) OFF — followed by the family's only automatic recall:

When engine start is completed, and APU BLEED pb-sw is deselected (automatic recall):

…at which point the line ENG BLEED (AFFECTED) — ON returns to the screen. ECAM kept the book open: when the reasons for closure (start in progress, APU supplying) expire, it re-presents the procedure so the bleed you hand-closed does not fly the whole sector forgotten. And the ground-ops connection: article 07's "wait for the APU BLEED memo before shutdown" exists precisely to deny this alert its favourite spurious trigger — a pneumatic valve commanded closed with no downstream muscle to close it.

HPV NOT OPEN. Trigger: This alert triggers when the HP valve is abnormally closed. — detected per article 05 by the transferred-pressure watchdog (<20 psig for 15 s with the HPV confirmed shut). Action: crew awareness. The STATUS line is the whole story:

NO BLEED 1 (2) AT LOW PWR

Article 02's three-case table explains the calm: the HPV only works cases 1 and 2 — low speed. In cruise at high power, IP feeds everything and a stuck-closed HPV changes nothing now. The debt comes due at idle: plan the descent knowing that side will go quiet at low power (ABNORM BLEED CONFIG will then coach the crossbleed). Some operators' MELs run the same logic in reverse — a stuck-closed HPV is dispatchable with the affected bleed simply not used at low power settings (article 11).

BMC FAULT. Trigger: This alert triggers when the BMC 1(2) is failed. The branch, from the FCOM procedure: APU bleed selected ON → ENG (AFFECTED) BLEED OFF; APU bleed selected OFF → ENG (AFFECTED) BLEED ON.

The reasoning is article 05's "three lost" list made operational. A dead BMC costs its side the FAULT light, the pylon-leak warning — and automatic bleed-valve closure. With APU air flowing, protocol demands the engine bleed yield (article 02's closure inventory); nobody is left to execute it, so your finger is now the BMC — select it OFF. No APU bleed, no yielding required — leave it ON and fly; the opposite computer's four-alert cover (article 05) watches over you.

[!warning]- The quietest alert with the biggest hole under it BMC FAULT sits in tier 1 of article 05's taxonomy — a single E/WD line, no chime, no light. Do not let the silence set the priority: from that moment, one side's automatic protection (auto-closure on overheat/overpressure/leak) is gone. The mismatch between annunciation volume and consequence is deliberate — nothing needs doing right now — but it belongs in your descent brief.

5. X BLEED FAULT and borrowing air (X OPEN)

X BLEED FAULT. Trigger: This alert triggers when the crossbleed valve position disagrees with the X-BLEED selector position. Action: X BLEED — MAN CTL, and the procedure hands you the AUTO logic's job description in one line:

Select OPEN, when the APU BLEED pb-sw is ON, or for engine start, or when WING ANTI-ICE pb-sw is ON with one bleed inoperative.

Three OPEN scenarios, otherwise closed — you are now the relay. Article 04's hardware makes this graceful: the manual motor is independent and essential-bus fed. If even manual opening fails with a bleed off, the closed-crossbleed wrap-up applies (article 08). STATUS records the mode (AIR X BLEED: MAN CTL) and, if the valve is stuck closed, the inoperative-system entry; the system-level version of this alert also folds in a wing-anti-ice consequence when the valve jams shut with one bleed out — anti-ice off, avoid icing — since a closed valve makes single-source dual-wing supply impossible.

ABNORM BLEED CONFIG (X BLEED OPEN) — the borrowing licence. Same trigger as the closed variant (at least one bleed failed, off, or not supplied). The five-condition audit, per FCOM PRO-ABN-AIR:

OPEN, if: ‐ No LEAK, and ‐ No ENG FIRE (detected, or FIRE pushbutton pressed), and ‐ No engine start valve failed open, and ‐ No overpressure with the bleed valve failed open, and ‐ No overtemperature with the bleed valve failed open.

Grammar as doctrine: the CLOSE list (article 08) is an OR — one contagion closes the door; the OPEN list is an AND — all five must be absent before you may share air. Borrowing is audited harder than refusing.

Once open, house rules — one source now feeds two wings' customers, so somebody economises: with wing anti-ice off and no engine failed, PACK FLOW — LO (Pack flow is limited to 80 %.) plus forward cargo cooling off; with wing anti-ice on or an engine failed, the affected side's pack — OFF (if the opposite pack is healthy). Two ways to shed load: everyone eats at 80 %, or one big customer leaves the table — and when anti-ice is involved, it is the pack that leaves (article 03's hierarchy: ice protection outranks comfort). The STATUS line condenses it: ONE PACK ONLY IF WAI ON. Run this configuration hot and long and section 2's HI TEMP is the natural sequel — the alerts of this family chain realistically.

A closing cross-chapter note ties bleed to the engines it comes from. The FCTM's engine-stall discussion lists A malfunction of the bleed system among stall causes — the mechanical reason behind article 07's ten-second rule — and its recovery guidance turns bleed into medicine: The flight crew selects the anti-ice on, in order to increase the bleed demand. This reduces the pressure at the exit of the compressor, and helps the airflow to circulate in the engine turbine from front to rear. The same extraction that can push a compressor toward surge can, deliberately increased, pull one away from it — bleed as both hazard and remedy, depending which side of the operating line you stand on.

Self-test

[!note]- Q1. State the first principle of this family and illustrate with BLEED FAULT versus WING LEAK.

Parameter sickness can be reset; physical sickness cannot. BLEED FAULT (a pressure/temperature excursion, often transient) earns one OFF-THEN-ON — OFF clears the latched autoclosure, ON retries. WING LEAK reports melted eutectic salt at a real hot spot: the isolation is latched by the continuing fault and no cycling reopens anything.

[!note]- Q2. Why does the BLEED NOT CLOSED sub-legend abolish the reset and import the leak playbook?

Because overtemperature with a valve that already ignored its electrical closure means un-stoppable hot air — functionally a leak. Hence straight OFF, APU BLEED OFF on the number-1 side, crossbleed CLOSED (condition five of the CLOSE list), and the users bricked up downstream.

[!note]- Q3. Where does 240 °C sit on the temperature ladder, and what is the complete list of things HI TEMP's procedure switches off?

Between 235 (pressure cut-back already active) and 257 (closure countdown not yet started) — the buffer zone. It switches off demand: pack flow to LO, cargo cooling to NORM, then wing anti-ice if persistent; with one bleed already out, the opposite pack. Never an ENG BLEED pushbutton.

[!note]- Q4. Why must autothrust be off for the LO TEMP push, and why does pushing work at all?

A/THR commands thrust symmetrically and would pull the advanced lever back; the fix is asymmetric — more speed on the cold engine only. Offtake temperature follows compressor speed, and the usual root cause is just idle power in very cold air, so added RPM often clears the alert outright.

[!note]- Q5. Explain both branches of BMC FAULT from the "three lost" list.

The dead computer's side has lost automatic bleed-valve closure. If APU bleed is ON, the affected engine bleed is supposed to auto-close and now cannot — the crew selects it OFF, finger replacing computer. If APU bleed is OFF there is nothing to yield to — leave it ON; monitoring continues via the opposite BMC's four-alert cover.

[!note]- Q6. HPV NOT OPEN: why merely "crew awareness", and what belongs in the descent brief?

At high power the IP stage supplies everything; a stuck-closed HPV costs nothing in cruise. The STATUS line NO BLEED 1(2) AT LOW PWR is the promissory note: at idle in descent that side loses supply, and ABNORM BLEED CONFIG will coach the crossbleed decision.

[!note]- Q7. Contrast the OPEN and CLOSE five-condition lists grammatically and philosophically.

CLOSE: five conditions joined by OR — any single contagion (leak, fire, start valve open, overpressure/overtemperature with valve stuck open) shuts the door. OPEN: the same five joined by AND as absences — all must be clear before air is shared. Borrowing is audited more strictly than refusing.

Key takeaways

References

Triggers, procedures, sub-legends, footnotes (FAULT-light extinction, the 80 % pack-flow limitation), the automatic-recall clause, the HPV STATUS line, the BMC FAULT branch, the X BLEED FAULT manual-control scenarios and the ABNORM BLEED CONFIG (X BLEED OPEN) conditions with their load-shedding lines and STATUS per FCOM PRO-ABN-AIR. Detection thresholds behind the triggers (257/270/290 °C ladder; 60 psig with 15 s closure / 25 s confirmation; the PRV LOW REGULATION table; the 20 psig/15 s HPV watchdog; dual-BMC coverage and losses) per AMM 36-21-00, AMM 36-11-00 and FCOM DSC-36-10-70 as developed in articles 02, 03 and 05; pushbutton reset semantics per FCOM DSC-36-20. Engine-stall causes and the anti-ice bleed-demand recovery technique per FCTM PR-AEP-ENG. The spectrum table and the "finger replaces the computer" framing are integrative syntheses of the referenced material. Dispatch notes reflect some operators' MEL practice (article 11).

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.