Pakistani Airbus A320 Crashed

[RVAmeche]

https://www.cnn.com/2020/05/22/asia/pakistan-plane-crash-intl/index.html

Approximately 100 on board, at least 11 fatalities so far, suspected engine failure

 

[Alistair_Heaton]

My mates in EK get additional manual hand flying sim sessions where they can practise what they like with a suggested list of things to do. They use them as much as they can on days off but shall we say they are unusual. But they do have to use days off and if they are doing ultra long haul they are banned from the sim centre 72 hours before departure. 24 hrs for normal flights. They can hand fly the aircraft below 10k but mostly don't because any complaints and its 4 days of hell in the head office with all the flight data pulled.

Its not the case that most pilots don't want to do the training or for that matter be the best they can but things are stacked up against them.

I have never had a issue with currency usually 4 sectors a day and if its visual the AP is out from as far away as I can with a turboprop. Never had a ride complaint from Cabin crew and no FDM violations so nobody cared.

This new machine I won't be doing it with until I get the feel of her. But then again I will be going to a lot more places where you end up number 6 on finals with 3 miles spacing and visual isn't an option.

The CS300/A220 has only one memory item which is put the O2 mask on everything else is either automatic or run through the EICAS. Q400 had 16 of them and JS had 20.

TOGA buttons on this thing leave the AP in which is a first for me. Hit the TOGA button and it pitches to 12 deg and brings the power up. Maintains runway track until 400ft then engages FMS and the only thing you have to do is put the gear up and flaps/slats but even they you can't get into to much problems because the AT targets 210 knts so the gear is safe and if you forget the flaps the blow back valves will protect them.

I don't have a clue how she handles yet I am told its different but nicer than the Q400. Even hardcore Boeing classic pilots who were extremely vocal about it being not Boeing and FBW shut up after they start flying it so my gut feel is its going to be quite nice and logical.

We aren't that far apart in our opinions about pilot training and what they should be able to do 3DDave. Its just after real life experience and knowledge how it all works I have realised that it will never happen to the entire pilot population.

My personal view with this is that it seems that the FEDEC and perm mag alternator is in a vulnerable spot, second is that AB needs a refresh of its flight control logic and system any way, third is what's the EGPWS getting its gear condition from.

That's the physical stuff, How to deal with people even attempting an approach like that I have no clue. Its been unacceptable for over 20 years but people are still doing it. And most of the time they will still do it and get away with it.

 

[LittleInch]

How does the FMS allow such a high and fast approach?

The updated blog I posted above has this for the final section which seems outside a"normal" approach. Now why the wheels weren't down first time around is not clear but excess speed doesn't look like it to me.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[EdStainless]

Juan Brown has put up 4 short videos about this. He is a commercial pilot (ex-USAF).

https://www.youtube.com/user/blancolirio/videos

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[Alistair_Heaton]

The FMS is a tool not the controller on a 19880's airbus.

It will not stop you embarrassing yourself or trying to kill yourself.

This whole thing started before the top of decent never mind 10k ft.

 

[LittleInch]

They are pretty good.

I didn't know until now that thet are three separate sets of marks on the runway from bouncing the aircraft on its engines.

Now the passengerr eye witness remarks, "he tried to land thrice " makes sense, I.E. the plane hit the ground three times before it managed to claw back into the sky. I realise it's probably instinct but must have been better to just scrape down the runway?

I'll get a screen grab of one of the graphs showing just how high and fast he was. It is frightening to see.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[3DDave]

They are still looking for the CVR. That's the primary interest as any major malfunction should have elicited a reaction from the flight crew and a request for special handling during the landing. It seems like everything but the flight crew was working properly until they damaged the engines.

This

https://www.pprune.org/rumours-news/632693-pia-a320-crash-karachi-40.html#post10795102

is an interesting observation. That the "gear down" warning is suppressed above 190 kts and the crew would ignore the terrain warning because they are landing.

So it's a software problem not issuing the correct warning on top of whatever else the pilots were doing.

 

[LittleInch]

Wow. You need to go back to post 724 for the initial post on the speed versus altitude warning envelope.

Also seems to inhibit below 159 knots and more than 200ft if the flaps are set but gear isn't or at least that's my interpretation.

That cvr is now crucial to see how this developed.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[LittleInch]

This is the diagram from video no 4 as posted by Ed. Blue line is our plane, redline is a "typical" A320 approach.

The only thing I can think of is that the pilot was expecting to enter a holding pattern for at least one sweep and due to lack of traffic got a straight shot in. But even so he seems to have just started descent from FL350 far too late.

The other graph showing vertical rate of descent.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Sparweb]

I've done one of those "drop from the sky" descents to land at an airport that cleared me to descent too late... once... in a Cessna 172.
I wouldn't ever consider doing that in an Airbus (surprised the machine allowed them to). That approach profile is so extreme, it is no wonder they couldn't get it stabilized before touch-down.

http://www.sparweb.ca

 

[Alistair_Heaton]

To be fair no pax onboard most turboprops would manage that with no issues. And actually I wouldn't be surprised if they had got the gear down they would have got away with it.

Until they got to about 3000ft there wouldn't have been any warnings about decent rate.

The problem you have with this sort of profile is the cabin pressurisation controller will not be able to keep the cabin even remotely pleasant. Anything more than about 750-1000 ft per min decent in the cabin from 8k will be extremely uncomfortable for the pax.

There are a few airports in Europe with other airports near by where you are held high and then most TP's go straight in and the Jets go off on a wee transition to get the height off.

There are actually a few ways of dealing with getting rid of height and it all depends on how much distance you have to do it in. The faster you are the less gradient you can produce. They teach on Jets that if you want to go down then you need to go as fast as possible. To create as much parasitic drag as possible. And its the most fuel efficient method. If you need a steep gradient your better chopping the power while flying level and then descending as slow as possible. At 20k or what ever your limit is throw the gear out and take it down. But doing that is extremely hard on your ears, you will catch the cabin at about 5000 ft and then the neg pressure valves will open up and the cabin will descend at what ever the aircraft is doing. I suspect they were trying to do that but put the gear out to fast and didn't realise. You get a huge lump of drag while the gear is in transit which most people use to wash off some speed so they can chuck some more flap out which then adds to the drag and everything is happy and stable.

Just to note some gear speeds for the A320

Vlo retraction is 220 knts
Vlo extension is 250 knts

Vle when the gear is down is 280 knts. There is nothing stopping you slowing to 250 knts putting the gear out then speeding back up to 280 knts.

Flaps/slats can start coming out at 230 knts.

Flaps/slats and gear max Level is 20 000ft asl.

So I presume they were at Vmo down to 20k and didn't come back to 250 knts to put the gear down. The handle was moved but they didn't spot it wasn't down. As I said before if they had managed to get the gear down I don't think we would be discussing this flight at all. it won't be the first slice in the cheese model but it could have stopped it.

TP empty you can be injected into an airport on the downwind at 18 000 ft and still get down and landed with no warnings off a 3 miles final Vmo at 4 miles and fully configured and stable at 500ft. As a pilot its very good fun to be honest. There is no way I would ever do it with normal pax in the aircraft. You couldn't do it in a piston aircraft because you would shock cool the engine.

A lot of company's have a hard limit of not more than 3000ft/min decent rate below 10 000ft for very good reason. I have never done such an approach in the Q400, we used to do them in the Jetstream with freight in the back. The engineers used to set the flight idle at as low as it would go legally so we had 4m2 of airbrakes on the wing when we went max rpm and flight idle. It would give a torque of -1 to 0% if they set it lower than that the automatic feather system would trigger. You would get 6000ft/min decent rate reducing to 3000 at 3000ft and 2000 at 2000 ft and max 1000 at 500ft and then no EGPWS alarms would sound. You could keep it at 6000ft/min but the EGPWS would start bitching and you would have a windshield full of ground.

 

[danschwind]

The problem you have with this sort of profile is the cabin pressurisation controller will not be able to keep the cabin even remotely pleasant

Yeah, my ears agree with you. When I was ~17 I thought of becoming a pilot but then I would probably be deaf already.

Daniel
Rio de Janeiro - Brazil

 

[Sparweb]

The problem you have with this sort of profile is the cabin pressurization controller...

Most planes have a button to help you with that:
DECOMP

There can be no hiding this cowboy maneuver from the ADS-B, AHRS, ACARS and probably a dozen other items to "snitch". If this plane was being flown in the western sphere with a western crew for an airline on a western OC, this kind of descent profile would result in what kind of disciplinary action against the flight crew? Retrain, demote, or fired?

http://www.sparweb.ca

 

[Alistair_Heaton]

Decompression switch there is usually 4 or 5 ways of doing it, either turn the bleeds off, shut the packs down, manually open the outflow valve and there is usually a separate emergency outflow that can be opened using via decompress switch its either attached to that or the main one The A220 has 2 emergency outflow valves powered off different busses. But if you do that your cabin pressure drops to what ever your alt your at, and the cabin follows then the aircraft profile with pressure change. If your above 13k the rubber jungle will come out and above 30k there will be people with decompression sickness onboard and also DVT. I have only ever done it once for fumes in the cockpit and when we did it we were below 10k and it was the forward valve to dump through the nose to clear the fumes. The cabin pressure didn't actually change at all. At that point we were 3 mins to landing so we stopped the colossal checklist and landed the aircraft. It turned out one of the radio transceivers had self destructed and was emitting a burning stink but still working. We were at 25k when it all started and it was pretty much Vmo with masks on directly to a 4 mile final coming back to 245 knts at 10k.

If your above 30k you more than likely will have the cabin go IMC with clouds forming in it.

But all you are doing is taking the control of the cabin pressure away from the cabin pressure controller.

The bit from 10k down to the runway would have me fired. I am not so sure my license could be removed though. The profile above that wouldn't.

In theory all ICAO countries should have a FDR analysis system in place for AoC holders so some 250 data streams will be recorded every 3 seconds for every flight. As per the MAX there are primary and secondary streams. The primary flight conditions and variables such as we are looking at here are set in stone to be recorded. The secondary streams are sometimes a bit woolly. And the company should download the QDR (quick data recorder) regularly and monitor compliance to regulations, SOP's and aircraft limitations.

Fundamentally the real screw up started when the gear was selected down above the aircraft protection window of 260knts and although the handle moved the gear didn't. As 3Ddave rightly said if there was an airspeed interlock on the handle it wouldn't have happened.

I haven't seen anything yet to say where the EGPWS gets its gear configuration from. If its microswitches on the handle then that's one major safety system gone.

The performance during decent must have been extremely strange without gear drag. Although for the vast majority of what you deem quiet rightly as a cowboy approach they were well above flap limiting speeds.

Anyway I don't think you will find any commercial pilot that will say that the decent profile is acceptable for a commercial pax carrying jet. From the parachute (meat bombers) pilots you will get black humour along the lines of I did 20 of them yesterday or simply "hold ma beer" they will quite happily come down faster than the meat bombs they have just delivered.

The Captain was a 17000 hour ex mil experienced guy. I really don't understand what his thought process was to attempt this. There are huge cultural differences between myself and an Asian ex mil pilot with double the experience I have. I can sort of understand what they were trying to do, I can understand the performance that they expected and can see that it might have worked if the gear had come down when they expected it. But the rest of it no clue. I wouldn't even attempted it, I would have just asked for more track miles. Once or twice round the hold would have easily sorted it.

The pilot screw up is a given. I am more interested in the safety systems and what information they were giving to the crew. All the machines I have flown so far there would have been warnings starting about 3000ft for sink rate then escalating with no way of muting them down to second stage of flap being selected. Once that was selected all hell would have let loose, gear horns, the EGPWS having a hissy fit. The eventual warning at 160ft would be the EGPWS shouting "too low gear". IF all the gear warnings had gone due to the handle being in the down position then its a concern and a major factor. The old Jetstream prior to EGPWS and TCAS being fitted ( yes I am that old and flew then) the alarms would still have started when the second stage of flap was selected and the gear wasn't down.

The other technical issue is the place meant of the perm mag generator and FEDEC on these engines. The A220 its up the back on the gear box and the underside of the engine is clean even of fuel lines. This is mainly because of the colossal bypass fans. This doesn't in anyway take anything away from the colossal pilot screw up I might add. It just shouldn't be possible for the pods to scrape the runway and it end up with a dual flame out 90 seconds later. This has now happened twice with this aircraft type. The first time in Tallinn they didn't have pax onboard and they managed to get it back on the deck nobody killed but the plane was a write off.

I mentioned it before as a human factor. Pilots are not meant to follow Ramadan while they are working. I have worked in Saudi during it and the local imams are more than happy to issue certificates stating that you shouldn't follow it as a pilot working. And the crew canteen is still open at the airport during the day. I got three certs one for my accommodation door, one for the car and one for the work bag. Now some Muslim pilots do follow it. 22nd of May was the second last day of fasting after it starting on the 23nd of April.

The plane crashed at 14:45 local time. Sunrise was at 05.44 if it was anything like Saudi with people following it the pilots will have been up since 04:00 eating/drinking with the family and then morning prayers. They will have gone to bed at about 22 or 23:00 the night before after Isha at 20:44 . Now its not outside a possibility that the whole of the crew on this aircraft had neither eaten or drunk anything for 10 hours in pretty oppressive high temps over 30 deg C. Personally I feel this could be a major human factor in this accident. I might add all of my Muslim colleagues over the years have been excellent operators and have always ensured they are fit to fly an aircraft and if working during Ramadan have either flown during the night shift or have broken their fast during the day to ensure flight safety. So from them I believe its perfectly possible to respect their religious obligations and be a safe pilot of a commercial aircraft. But we shall have to see if its a factor in the accident. My gut feel is that it might be a major factor. Dhuhr was at 12:30 ish so its not outside the realms of possibility they missed top of decent due prayers. I have landed at lunch time prayer times with no ATC when it should have been active. All the pax have got off and promptly started prayers on the apron and nobody had an issue with it. In the western world there would have been a hissy fit by the regulator and by apron safety ops. First time I went with an experienced in area FO and just landed and then hung around until they went inside. After that I just went with him inside as it was ops normal for the region and absolutely nobody thought I was strange or had an issue with it and I really didn't want to hang around in 40 deg C.

 

[LittleInch]

The other system issue mentioned on pprune is that the alarms don't actually go off above certain air speeds. The designers of the system clearly assumed no one would try to land at over 200 knots.

If I recall the AF plane they found out the stall warning stopped going off once you went above a certain AoA. Hence when they tried to put the nose down the stall warning kicked in and confused the hell out of them. Only one of many things they did wrong of course but why some of these limits apply isn't always very clear.

No gear down above a certain speed is a physical issue but no warings above a certain air speed.

As an ex "meat bomber" myself I think it was a matter of pride to the pilots that the plane would land before the ex pax. Average of 4000 ft/min.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Alistair_Heaton]

I know the various EGPWS modes but have zero clue about how the airbus logic works.

This gear protection envelope is on my list to be looked up in the FCOM and if I can't find anything then ask the sim instructor on the A220

The Q400 if you put the handle down the gear is coming down what ever the speed is even if it rips the doors off same as the other TP's

Its got more than a few slices of cheese lined up this one from both a human factors and technical POV this one. I am sure it will feature in many a CRM/TEM recurrent course for many years to come.

 

[Alistair_Heaton]

Had a grill yesterday with people that have worked in the region.

The consensus was dehydration screwing with thier heads.

Must of us seem to go through half a ltr of water an hour in the cruise. And working in those temps we would be drinking a ltr or however much is required to go for a pee every 2 hours. It was a surprising consistent volume across age range 25 through to 60 male and female.

By 12.45 and drink at 5:40 I would have put away nearly 7 LTRs of water and one of those electrolight dehydration saches, An omelette and a bucket of chicken salad. If I wasn't working it would be 5 LTRs of water, dehydration sachet and chicken grill with humous and Arab bread.

I don't think they mentally 100%

 

[hokie66]

Interesting theory, Alistair. And it was near the end of Ramadan. Fasting?

 

[Alistair_Heaton]

Second last day of Ramadan. It started on 23rd of April. I have seen nothing to indicate either way if they were fasting or not.

All it is, is a theory for a human factor.Gut feel is it is.

The tech issues with Airbus logic, warnings and engine design still need looked at, what ever the human factors are deemed relevant.

I don't know much about Pia to be honest. Never flown with anyone that's flown for them. It's not on my no fly list yet.

I don't know if they will release the CVR which would hopefully indicate if there was a mental state issue.

Blood sugar and dehydration is a well known feature of being a pilot. It's the individuals responsibility to ensure our performance is not compromised by it. Any incident can murder the blood sugar level taking it very low. Until the secondary effect of adrenaline opens the taps up from fat stores. Some off us carry a couple of packs of diabetes dextrose sweets in our bags. They usually get used up with sick pax and cabin crew. But if you get a divert after not getting the runway they are handy when you have been awake all night.

 

[LittleInch]

The CVr and FDR have both apparently been recovered and sent to the BEA in Paris via a special Airbus plane.

PIA requirements for flight crew and fasting are not clear - some say it wasn't allowed, others that it was in practice.

There are some mentions of a "false glide slope" but there's very little out there as to why any of this happened.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[EdStainless]

I would buy the glide slope issue if they were coming in at 6deg instead of 3deg, but they were obviously on the edge of having any control.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed