Caverton AW139 incident in Jan 2024

[nowherespecial]

Summary: Loss of AP while airborne leads to unusual attitudes for several minutes in flight. Many unhappy passengers.

This looks like it was a bit exciting. Anyone have any insight as to if the correct procedures were followed?

https://nsib.gov.ng/wp-content/uploads/2024/03/Preliminary-report_5N-BSG-1.pdf

[Nescafe]

The QRH states that following a single AP Fail

”- If subsequent 1-2 AP fail caution illuminates
- Continue flight manually not exceeding 140 KIAS (100 KIAS in turbulence,
IMC or approach)”

Nowhere does it mention resetting them repeatedly or shutting off the gang bar.

[torqueshow]

Anecdotally the 139 year would regularly drop APs, especially on approach and it wasn’t uncommon for the PM to “guard” the AP buttons on the AFCS panel for a quick attempted reengagement.

Nothing wrong with seeing if it will come back on. But 5 attempted resets followed by turning off both generators at the same time in flight via the gang bar is bizarre.

The 139 can be a handful without the APs but not 3000ft to 300ft in 45s, perhaps more time should be spent on training manual control of the aircraft with and without AP. Seems like subsequent loss of AP1 after messing around with the electrics wasn’t expected followed by panicked over controlling.

[nowherespecial]

Megan, I think Nescafe was referring to the QRH not mentioning resetting the Gangbar. The report clearly states that's what the crew did.

Thanks all for the information. Sounds like a major mishandling of the aircraft then.

[griffothefog]

It’s always ok when everything works, but simple failures offered in the sim result in the most bizarre behaviours in my experience ( as a former TRE on the 139 ), some would have me tightening the seatbelt even… 🤭

[deeceethree]

The Preliminary Report makes disturbing reading. The two pilots seem utterly preoccupied in keeping the autopilots engaged, and repeated pressing of buttons seems to be their preferred course of action, during which the aircraft is all over the place:
According to FDR data, simultaneous with the attempts to re-engage the AFCS, the aircraft climbed to about 3,500 ft within a period of 1 min and 40 s, followed by a rapid descent to 330 ft in 45 s. FDR shows that the aircraft commenced another climb to 2,500 ft in the next 2 mins within which period, several pitch and roll control inputs were recorded; pitch angle varied between 28° pitch up and 12o pitch down before a maximum 25° pitch down motion was recorded, simultaneous with the maximum recorded roll angle of 37° to the right.

Over the next 44 s, the PF attempted to correct the right roll unto datum and continued into a left roll to a maximum angle of 15° before recovering to level flight 9 min later, following a series of roll attitude corrections.


The 'Initial Findings' section of the Prelim report states "8. The serious incident occurred in Instrument Meteorological Conditions (IMC)." The preoccupation with autopilot engagement (never mind the 45 second dive from 3500 ft to 330 feet!) leaves me with a distinct impression that the pilots were very reluctant to manually fly the aircraft! Is that because they are not confident about doing so in IMC, perhaps rarely practising doing so?

A few more items of interest from 'Initial Findings' section of the Prelim Report:
3. The First Officer’s medical certificate validity expired on 15th January 2024.
[Expired 5 days before this incident, which occurred 20th January 2024.]
......
9. The crew experienced failure of the Auto Pilot systems (AP 1 and AP 2) and the
aircraft entered ’unusual attitude’.
......
10. The FDR captured multiple unusual attitudes within a 5-min period.
......
16. The CVR recordings of the event were overwritten."


Additionally, "The aircraft sustained minor damage.":
The following damages were observed during post occurrence inspection of the aircraft:
1. Two cabin windows were blown out in flight, and missing on ground.
2. Debris from a broken third window was found in the cabin.
3. A main rotor blade had a broken glass shrapnel stuck to the blade root.
4. The main rotor lightning conductor was broken.
5. A cabin light lens and transparent light cover were broken.


Many unhappy passengers.
Yes, I can sympathise with that. 😲

[212man]

I am in no way defending the crew's actions but, when discussing manual flying, I think it's worth remembering that having turned off the generators they were left with standby instruments only and zero stabilisation (in IMC), so it was a bit more complicated than simply hand flying. Nonetheless, pretty astonishing to have been out of control for that length of time. I find the crew's actions in making non-standard interventions, resulting in a loss of control, reminiscent of the AirAsia accident a few years ago: https://en.wikipedia.org/wiki/Indonesia_AirAsia_Flight_8501
n December 2015, the Indonesian National Transportation Safety Committee (https://en.wikipedia.org/wiki/National_Transportation_Safety_Committee) (KNKT or NTSC) released a report concluding that a non-critical malfunction in the rudder (https://en.wikipedia.org/wiki/Aircraft_rudder) control system prompted the captain (https://en.wikipedia.org/wiki/Pilot_in_command) to perform a non-standard reset of the on-board flight control computers (https://en.wikipedia.org/wiki/Fly-by-wire). Control of the aircraft was subsequently lost, resulting in a stall (https://en.wikipedia.org/wiki/Aerodynamic_stall) and uncontrolled descent into the sea. Miscommunication between the two pilots was cited as a contributing factor.[1] (https://en.wikipedia.org/wiki/Indonesia_AirAsia_Flight_8501#cite_note-Karmini-2)[2] (https://en.wikipedia.org/wiki/Indonesia_AirAsia_Flight_8501#cite_note-3)[3] (https://en.wikipedia.org/wiki/Indonesia_AirAsia_Flight_8501#cite_note-4)

It will be very interesting to see the final report and more data on the flight trajectory. Anecdotally I have heard they fell out of the cloud base at 1500 ft doing 13,000 ft/min (7 seconds to impact) and pulled 4g in the recovery!

EDIT - I am hearing from a friend that when you move the electrical gangbar it takes out the battery too, so also the standby instruments (an integrated EFIS). So, even when you realise your error and reset it, everything has to power up, go through their BITs etc and come on line. So, unstabilised in IMC with no instruments - sporty!

Can any 139 pilots confirm?

[malabo]

The 139 was developed and marketed as a fully PC1 aircraft, including procedures and profiles to deal with that dreaded single engine failure.
On approach, the profile will settle you in a high vibration regime that could kick the AP's off. If you'd flown a 500 and instinctively minimized time in paint-shaker mode you'd be up in front of the FDM board explaining why you deviated from the PC1 profile.
I don't know Caverton's training, but likely focussed on engine failures (PC1 myopia). Not much time left over for AP's off, or standby instrument recovery from IMC. Not much different in the rest of the world.
I flew for a large operator that would regularly do both AP off training in the 139 aircraft, as did the factory. I noted other operators prohibited their training staff from doing this, citing a POH limitation.
Nothing like lifting off a deck at night with both AP's still off ....

I recall another highly-feted Nigerian crew ditching an S76 offshore after forgetting the AP's and deeming the aircraft too unstable to fly. Limited training time, focus on what the regulator wants to prioritize. In another case a droopy-eyed expat lost it in cloud between the beach and PH, saved by his sharp Nigerian national copilot.

[212man]

I recall another highly-feted Nigerian crew ditching an S76 offshore after forgetting the AP's and deeming the aircraft too unstable to fly.
Not quite - left the compasses in DG then the AP didn’t behave as they expected. But, hailed as heroes in the local press

or standby instrument recovery from IMC.

Seems like they may not have had any instruments for a bit - see my last post.

[Nescafe]

​​EDIT - I am hearing from a friend that when you move the electrical gangbar it takes out the battery too, so also the standby instruments (an integrated EFIS). ​​ ​​​

From the AW139 Technical Notes

”Red Gang Bar moved backwards cuts off all aircraft electrical power sources at the same time. (Gen 1 & 2 and Battery Master switches are moved to OFF all together.)”

The Battery Master switch needs to be on to enable the Main and Aux to be connected to Main Bus 1 & 2.

[212man]

From the AW139 Technical Notes

”Red Gang Bar moved backwards cuts off all aircraft electrical power sources at the same time. (Gen 1 & 2 and Battery Master switches are moved to OFF all together.)”

The Battery Master switch needs to be on to enable the Main and Aux to be connected to Main Bus 1 & 2.
Thanks - seems to corroborate my info. So, given that most AHRS normally take about 30” to come online when sitting on the ground motionless, and a minute or two in flight, or on a moving deck, one can imagine that they take a bit longer while doing aerobatics!

[casper64]

Just FLY the damn chopper! Or have we forgotten how???

[212man]

Just FLY the damn chopper! Or have we forgotten how???
I don't think anybody would argue with that sentiment, but it's not clear what phase of this event you're referring to? With the AP drop out, then of course fly the aircraft and follow the QRH - which says decouple the Flight Director, ergo fly manually with the remaining AP in ATT mode. Some OEMs say try to attain VMC, but I see the 139 QRH does not say this. It does say that if the other AP fails then reduce speed to below 100 KIAS, and it also says (in the limitations section) that taking the APs out of ATT mode in IMC is forbidden.

However, from what we are now learning, the captain used the electrical gangbar, which turned off all the electrics at once, leaving them with no flight instruments at all, and a totally unstabilised aircraft which, I am sure, would have rapidly departed normal flight. Even with the electrics restored immediately, there would not have been any usable attitude information for quite some time, so the concept of "just flying the damn aircraft" is a bit moot at this point - they were essentially passengers. I think that had the cloud base (which gave them the visual references for recovery) been much lower this would have been a fatal accident.

Caverton has a Level D FFS in their main base, so it would be very interesting to see how this event looks like replicated in there! No doubt, that has already happened....

[Uplinker]

A graphic example of why it is a very bad idea to make up your own drills or actions. Thank goodness that all survived this, with only minor damage to their underwear.

This gang-bar reset thing sounds as if the Captain had done it before, but didn't think about the battery also going off and losing the standby instruments - I mean who would intentionally switch off ALL electrical sources in flight, unless the actual published QRH or published memory drill demanded it ?

Out of interest; what caused the windows to smash in this incidence ?

[Aser]

With the AP drop out, then of course fly the aircraft and follow the QRH - which says decouple the Flight Director, If during Non SAR mission
- Continue flight attentive do not exceed Vne-27 KIAS in level flight, or 100 KIAS and 1000 fpm in climb.
- Decouple FD (below 500 ft (152 m) AGL fly manually) unless in approach, missed approach,AFCS approach and departure to/from hover and hover

As another pprune user usually says, just start the clock and don't touch anything... :}

[212man]

If during Non SAR mission
- Continue flight attentive do not exceed Vne-27 KIAS in level flight, or 100 KIAS and 1000 fpm in climb.
- Decouple FD (below 500 ft (152 m) AGL fly manually) unless in approach, missed approach,AFCS approach and departure to/from hover and hover

As another pprune user usually says, just start the clock and don't touch anything... :}
I have an old uncontrolled copy that has similar but slightly different wording - thanks for the update.

Out of interest; what caused the windows to smash in this incidence ?

​​​​​​​My assumption is either fuselage flexing, or extreme side-winds, but just a guess.

[Bravo73]

Out of interest; what caused the windows to smash in this incidence ?

It could have been the passengers, looking for a way out.

Wouldn't you consider it if you were down the back?

[Uplinker]

To get out on the ground, possibly if the door didn't open and getting out was imperative. But there was no reason to exit in an emergency, was there ?

For some reason, I thought they meant the windows had smashed in flight somehow.

[Captain Catastrophy]

Interestingly this recently cropped up in the sim ( unrelated ) and - as stated - the gangbar shuts down everything and takes about 30 seconds for everything to reboot during which time the aircraft could do anything. In this case the aircraft was recoverable but was in a very unusual attitude when the lights came back on.............. Turning everything off ON THE GROUND was an unofficial engineering CTR ALT DEL to clear snags otherwise requirering a shutdown etc etc. It was never recommended to crews and certainly not in flight...

[212man]

Nagging question for 139 experts - how was the FDR recording the flight parameters after the power was cut? I can imagine it was powered up immediately after the gangbar use, but what about the data?

[Torquetalk]

Nagging question for 139 experts - how was the FDR recording the flight parameters after the power was cut? I can imagine it was powered up immediately after the gangbar use, but what about the data?

Not sure this counts as “expertise“ 212man, but the gang bar cuts all the electrics for pretty much everything that the pilots care about, except the engines. But the aux batt is still supplying power to the CMC and recording data until physically disconnected. It‘s just the BATT BUS that is offline.

The overwrite of the CVR is a gem. Malpractice by both pilots and managers in this regard needs some serious attention and penalties by authorities, including those well North of Nigeria.

I think the referred to IMC conditions means something less than VMC and short of being in cloud. Or they would all be dead.

[212man]

Not sure this counts as “expertise“ 212man, but the gang bar cuts all the electrics for pretty much everything that the pilots care about, except the engines. But the aux batt is still supplying power to the CMC and recording data until physically disconnected. It‘s just the BATT BUS that is offline.

The overwrite of the CVR is a gem. Malpractice by both pilots and managers in this regard needs some serious attention and penalties by authorities, including those well North of Nigeria.

I think the referred to IMC conditions means something less than VMC and short of being in cloud. Or they would all be dead.
Thanks, but I was trying to understand how the FDR would receive AHRS data if they had been powered off and were rebooting when power came back?

[sycamore]

There used to be a `get -out-of -jail card`,called a standby AI/AH.Usually a 4"diameter representation of the outside world and a 3 position switch annotated `ON`/off/EMERG`,powered by the battery,and a small emergency battery,in case the `battery` failed,after everything else.A length of coloured cord on the windscreen also helped..Cost...next to nothing in a million-$ helo...Don`t they have those any more...???

[Collective Bias]

The aircraft could have RIPS. Standard after a certain s/n. Remote Independant Power Supply just for the FDR/CVR. Can’t remember when it became std, but before s/n 500.

CB

[212man]

The aircraft could have RIPS. Standard after a certain s/n. Remote Independant Power Supply just for the FDR/CVR. Can’t remember when it became std, but before s/n 500.

CB
I can understand the FDR receiving power, but how is it getting data from the AHRS etc if they were off then realigning?

[Torquetalk]

I can understand the FDR receiving power, but how is it getting data from the AHRS etc if they were off then realigning?

No doubt some significant interrupts in the data. This would compliment the lack of data from the CVR. Whoops. How could such a thing happen? Again. And again.

[ericferret]

[QUOTE=Collective Bias;11621700]The aircraft could have RIPS. Standard after a certain s/n. Remote Independant Power Supply just for the FDR/CVR. Can’t remember when it became std, but before s/n 500.
microphone (

This is from the AAIB accident report to G-LBAL S/N 31421
"The CVFDR was fitted with Recorder Independent Power Supply (RIPS) designed to keep
the audio recording of the cockpit area microphone working for 10 minutes after the loss
of the main source of electrical power to the CVFDR. "

[the coyote]

Hopefully the use of the gangbar is just a reporting error, as I am astounded that the Captain (or any half competent pilot) would do this in IMC and remove all electrical power, and thus all instruments.
That is suicide in IMC and reflects no system knowledge and dangerous training deficiencies if it is actually the case.
I agree with TorqueTalk, they must have had some sort of visual reference or they would have been dead.

The ONLY time to use the Gangbar is on the ground.

[Nescafe]

The ONLY time to use the Gangbar is on the ground.

Agreed, but if the crew have become accustomed to relying on it to sort glitches on the ground (bearing in mind that it is NOT an approved remedy) it can be a reflex action to reach up and flick the switches. In this case, instantly regretting your actions.

[212man]

Agreed, but if the crew have become accustomed to relying on it to sort glitches on the ground (bearing in mind that it is NOT an approved remedy) it can be a reflex action to reach up and flick the switches. In this case, instantly regretting your actions.

I refer you to my earlier reference to the AirAsia accident. Plus, the anecdotal information that they pulled 4g recovering - which points to being fully IMC until they came out the cloud bottoms.

[Torquetalk]

Agreed, but if the crew have become accustomed to relying on it to sort glitches on the ground (bearing in mind that it is NOT an approved remedy) it can be a reflex action to reach up and flick the switches. In this case, instantly regretting your actions.

Spot on

but how does that “reflex” happen in flight without crew coordination?

I got this - oops
You don’t exist - oops
Commander knows best - oops
Let’s do something - oops

Was this really a simple non-critical malfunction that nearly killed everyone due to crew ineptness? Amazing and dismal in equal proportions.

And they have a company simulator? So what do they train? The things that the regs require but which are rarely, if ever implicated in helicopter PT cock-ups?

[Nescafe]

And they have a company simulator? So what do they train? The things that the regs require but which are rarely, if ever implicated in helicopter PT cock-ups?

Someone mentioned earlier “PC1 Myopia.” How true.

Most half competent crews can manage engine failures before and after TDP and they can work their way through a fire drill, but throw them an AHRS fail or a niggling ADS malfunction and then you’ll sort the wheat from the chaff.
Personally, I believe we aren’t getting the balance right.

[Scorpygixxer]

A graphic example of why it is a very bad idea to make up your own drills or actions. Thank goodness that all survived this, with only minor damage to their underwear.

Out of interest; what caused the windows to smash in this incidence ?
Passenger probably hit it hard enough to break it during the PIO

[Scorpygixxer]

Having seen too many of these type of incidents recently, my attitude to the traditional sim 'let the PM initiate the UA and the PF recover' has changed. I now initiate the UA sequence with the APs off and challenge the PF to remain adequately stabilised for 10 seconds before switching the APs back on. Some guys almost immediately enter their own UA.If they are good, I will ask them to look at the AP panel for 10 seconds or close their eyes until it occurs. Then ask the PM to reselect both APs and the PF can recover. I also repeat the ex in SAS mode without ATT to recover. Everyone forgets to use their feet! This tends to remind crews of the importance of a quick scan and quick crew actions to get the APs back ASAP after getting it 'back in the ballpark'.

Switching off the gang bar in IMC is tantamount to attempted suicide. No SAS, no instruments, no ICS, no AWG, no chance. AHRS can take up to 60 seconds to realign and considering the rates of likely pilot input, probably longer.

[MaxiOne]

For me there is an element of doubt about the whole thing. I know the 139 has AP issues and that is fair enough but when you look at the timings of the incident and then how long they spent on the ground it just doesn't sound like normal behaviour. By staying on the ground with the CVR on for 47 minutes it allows it to restart the 2 hour cycle and wipe all voice recordings of the incident.

Maybe something in it, maybe not, maybe just the musings of a Southern North Sea jockey.

[Nescafe]

It could be a very simple oversight. Aircraft gets back, engineers plug in external power to start analysis and don’t pull the FDR circuit breaker.

[MaxiOne]

It could be a very simple oversight. Aircraft gets back, engineers plug in external power to start analysis and don’t pull the FDR circuit breaker.

That would be the case if they had lost all the data but the CVR had recorded two hours worth of data. The long delay on the ground caused the loop to start again omitting the incident.

[MaxiOne]

That would be plausible if the whole thing was wiped but the incident happened about 2 hours 10 minutes before the power was shut off, the crew sat on the ground for 47 minutes. The CVR records on a 2 hour loop and by waiting 47 minutes then it gives about 10 minutes allowance to make sure it was wiped. I think there is more to this than meets the eye.

[albatross]

It could be a very simple oversight. Aircraft gets back, engineers plug in external power to start analysis and don’t pull the FDR circuit breaker.

Shouldn’t there be a checklist that engineers / investigators should have to follow to preclude such an error before plugging power into the aircraft.
Imagine putting electrical power into a wreck, causing a short and having residual fuel or wiring harnesses etc.burst into flame not to mention various computers and screens having a meltdown or a wx radar start emitting.

[Torquetalk]

If the aircraft has had a significant safety event, all parties involved should be quite cognisant that data needs to be preserved. When that doesn’t happen, it smells mighty fishy right away.

Two examples that come to mind involved the crew erasing the CVR in one instance and a corrupt manager ensuring overwrite in another.

It happens. And it happens in organisations with all the right bits of paper and post holders. None of it is worth a fart if the players have poor ethics and there is weak oversight.