What happened?

On September 10th, 2023, an Air China A320NEO carried out an emergency landing into WSSS/Singapore Changi international airport after the flight deck and cabin filled with smoke.

For those interested, the cause of the smoke was a carbon seal in the No.3 bearing compartment which had fractured. Oil leaked through it into the hot areas towards the back of the engine and turned into gross oily fumes and smoke which got sucked into the bleed air and, subsequently, into the cabin/flight deck/peoples’ mouths.

What I think is worth talking about today is some of the other things that occurred during this event. Namely, the communications, because they were good and not so good, and in it all there is a whole bunch of stuff we can all learn from.

Before we dive in, if you want to read the full report then you can find it here.

The Good.

Right, there are a few things to talk about here. Let’s start with the cockpit cabin comms because these were pretty darn decent.

How it went down.

The flight deck noticed a stinky smell, and the cabin crew also called to say they could smell something. The flight deck crew however decided it was to do with outside air, compounded by the fact hey had no ECAM alerts suggesting issues in the air systems (or any others that could be causing the smell). When the smell intensified and the cabin called back and reported smoke, the flight deck popped their oxygen masks on and declared a ‘Mayday’.

Pretty decent comms throughout!

Points worth pointing out.

Comms between the cockpit and cabin are super important in any situation, but it can be difficult to establish the severity of a situation because often reports can be subjective. The same goes for levels of smoke. What does “thick” mean to you? What does “smelly” mean? Is it coming from a particularly stinky passenger, or actually from the aircraft?

So when smoke or fumes are present, getting updates to help assess and review the situation is important, but so is ensuring what you say/are told is useful (and non-subjective).

Asking questions like the following can help:

• How many rows are reporting they can smell the odour?
• How rows can you see?
• What colour is the mist/smoke?
• Are there other signs or symptoms?
• What does it smell like?

In general, a smell like wet socks, or other moldly, swampy stinks generally equal oil in the air supply and that is never a good thing. Acrid, chemical smells can come from a myriad of sources (including passengers!) so signs that it could be an aircraft system causing it would be stuff like if smoke, vapour or mist are identified, the smell is widespread, or there are other signs such as skin or eye irritation.

IFALPA offer some good guidance here on all of that and its worth a read/share with crew.

The cabin really can be the eyes, ears (and nose) of the cabin.

In 2014, an Emirates A332 evacuated in Karachi after a hydraulic pipe burst, the fluid was heated and leaked into the cabin as a foul mist. This is a nasty fluid that causes a lot of eye irritation. An interesting feature of this event was that the cabin experienced a far worse situation than the flight deck, leading to them effectively telling the flight deck an evacuation was needed.

Read about it here on AvHerald if you want to.

Right, getting completely off track, but the point hopefully highlighted here is that establishing good comms between flight deck and cabin is critical. Crew need to feel comfortable speaking to the flight deck if they see or smell things out of the ordinary, and clear guidance on what to report and questions to ask can really help.

The less good (but great to learn from)

The communications between the flight deck and ATC, and between different bits of ATC were less good. This is not intended as a criticism of them, but to highlight how comms can break down a bit, and what we might all want to think about more during an event like this (if we are unlucky enough to ever experience it).

How it went down.

The crew declared a “Mayday”, informed ATC they had smoke in the cockpit and requested priority landing due to the emergency. On transfer to ‘Inner Approach Control’, they were advised to expect an RNP approach for landing runway 20L. The crew requested landing 02R but this was not accepted, and they were cleared 20L. The crew then requested 20R (possibly what they had meant to ask for before) and were told to “standby”. Under the airport’s guidelines (guidelines, not rule!) runway 20L was generally preferred for emergencies, and the tower controller wanted confirmation the crew could not fly an RNP before 20R could be allocated.

During the approach and landing roll, the crew advised that they would be stoping on the runway. There was a fair amount of back and forth between the crew and ATC about whether they would stop on the runway or be able to vacate. A lot of this took place during the landing roll and while the crew were attempting to establish what the situation in the cabin was like. At one point the crew said “We cannot evac… vacate the runway” which possibly confused tower.

There was also a bit of a breakdown in comms between ATC and the ground emergency services. In short, the ‘MAYDAY’ was not passed on, so the emergency services didn’t raise it to the higher severity level and were just on a ‘local standby’ level of readiness and response (which meant a bit of a delay).

Points worth pointing out.

Pilots are human and make mistakes, especially when dealing with stressful situations (like your eyes and nose filling up with oily fumes). So they might not make the greatest of communications – like mixing up runways, or being less clear on what they need. ATC need to be aware of this and try their besty best to support crew, and they can do that by:

• Reconfirming requests if not clear
• Waiting for appropriate times to request information
• Going pilots what they ask for (unless there is a safety reason not to)
• Understanding there might be delays in crew responding, particularly if they are trying to work things out or run checklists

Pilots on the other hand can do their bit to help ATC help them by:

• Being super clear with what they need
• Sticking to brief and concise communications (and standard communications)
• Understanding that misconceptions and misunderstandings might occur and helping avoid these (by doing the above)

A particularly major point is that there is often a lack of awareness and understanding between roles as to the challenges the other might face. For example, ATC requesting crew fly an RNP approach to a different runway means:

• An increased workload as the aircraft has to be programmed and set up for this
• Performance might need to be recalculated
• The pilots will need to rebrief
• An ILS is often ‘easier’ just because it is what we tend to be more used to, but in something like a smoke situation we can also plan an autoland off it (very useful if we can’t see much in the flight deck)
• Distractions (questions at inopportune moments) break the train of thought/checklist and can result in more mistakes and/or delays from pilots.

There was another issue within this event worth talking about: the emergency does not end just because you’re on the ground. For pilots, decisions over whether to evacuate or not remain, and other hazards pop up (like ground vehicles zooming about potentially running over passengers who are scurrying about in a panic).

Thankfully, this didn’t happen in this event, but because the ‘Mayday’ had not been passed between teams, there were some delays and lack of coordination and awareness. The ground teams are often a critical information provider to flight crew regarding fire situations) which they were in this. So, what can we do:

• Ensure the ‘MAYDAY’ is passed between teams. ATC and crew can do this by simply stating it, and ensuring it is acknowledged with each transmission
• Keep comms clear and precise between everyone. Pilots use the NITS briefing and that can actually be used between anyone and everyone.

The things we can learn from it all.

Comms are key, and that doesn’t just mean what is communicated but how and when. So, clear and brief, standard, and not when it might be a distraction to more important things going on.

Understanding the situation and the challenges between roles, and having an awareness of the stress levels and workload of other roles is also critical, as is noticing tone and potential errors and realising that that might mean someone is at capacity, overwhelmed, or needing time to work through a situation before they can properly respond.

For crew, a big takeaway from this event should be how telling ATC (and I don’t mean in a rude way, but in a firm and decisive way) what you need is really important. They might have misconceptions or may not have the full information, so sharing the old mental model and making your “perspective clear from the beginning” (to quote the report) is important. Actually, you know what, same goes from ATC in a whole bunch of situations.

Communications and important, communicating well is critical, no matter what the situation.