While CRM focused on the individual and checkrides focused on individual skills back then, the current focus is based on the acceptance that no matter what, people will make mistakes, and instead rely on layers of safety nets to catch and stop a threat before an undesired state occurs

 

Still going on about this :)  Yes you are correct, safety nets are put in place, one of which is called stable approach criteria! Something everyone on the flight deck chose to ignore. 

 

Just out of curiosity, how many crashes have occurred at SFO on approach? As far as I am aware 2 in the last 40 years, the previous crash was with a functioning ILS and occurred November 22, 1968  

 

I agree with what you are trying to say, however it is like using a sledge hammer to crack a nut.  We are talking about millions of hours worth of safe operation over the years, now 3 guy's screw up and fail to even follow one of the most basic SOP's (Stable by 1000ft IMC/500VMC)  and now you suggest the whole system is dangerous and should be thrown out of the window? 

I don't know how many crashes were at SFO. I am pretty sure there have been more than a few in the history of aviation that can be attributed in some way to an unstable approach. How many more crashes at SFO would be necessary for you to feel that maybe controllers there shouldn't be holding planes high?

 

What is so wrong to you about doing something about an identified threat to safety? I'm not asking for the entire ATC system to be thrown out. All I'm asking for is a descent clearance 30 seconds earlier and for the controller to kerp the guy behind me far enough back that I can configure and slow the plane according to how I was trained to do it and how the plane was designed to be flown instead of having to hold onto a speed until the last second and then throw everything out at the limits and hope the tail wind isn't too much for me to get things stabilized before I have to go around. Because you know just how dangerous a go around is. Is that too much to ask?

All the foregoing, of which, suggests that we may be reaching the limits of aviation growth - unless new technologies emerge to solve the conflict between expanding people populations and ever growing numbers of larger commercial airplanes.

january

 

 

 

What is so wrong to you about doing something about an identified threat to safety?

 

Hey Kevin,

 

Once again I do agree with a lot of what you are saying in general. In the context of Asiana I just don't feel the role of ATC is the big issue here. I don't know if you fly into KSFO? I also don't know anyone that operates into KSFO shorthaul on a regular basis. would be good to hear some more opinions. 

 

  I do know people that operate ULH into KSFO and don't see anything unsafe about the way things are done. These are flight crew that  operate half way around the planet and still manage to land safely.  I have often heard the term slam dunk used, have also talked to guy's that mention being held high (not just at KSFO)  Never have I heard the term 'Threat' or 'dangerous' used, challenging yes but never unsafe.  

 

You have Lufthansa operating A380's, BA, Emirates, Singapore & many others safely operating into KSFO.   

 

The arrival procedures used are within the capabilities of the aircraft and the crews if correctly trained.  The only issue I see is with this particular crews training.  Anyway, Just my opinion :) It's interesting to read others take on this accident.

 

Cheers

ATC was just one of many issues here. I pointed it out here because I noticed the NTSB avoided it even though it was obvious to me and everybody else I fly with and have discussed this crash with. The big issue here is still the crew being complete azzes and elbows trying to get that plane down and slowed downed. The controller only gave the ball the first small nudge towards the cliff by putting them in that situation. But it should have been addressed by the NTSB nonetheless. Theirs is not to placate the business interests or to preserve fault for a chosen party to aid litigation or prosecution, but to call out each and every safety issue affecting an accident and then let the chips fall where they may.

If it is a standard procedure at KSFO then crews should prepare for it. However, this seems to be another case of too much reliance on the a/c systems and lack of airmanship. What ever the angle of approach, a good landing is made at the top of the glideslope. The crew must be ahead of the a/c all the time. Such that any deviation from the expected is countered before it happens. So one would expect that for a "non" standard approach, the monitoring would be even more intensive. The one thing you don't do though is allow the engines to spool down to idle when very low. So if the a/c systems are demanding something, that to you is contrary to basic airmanship then you must intervene.

Look at a couple of examples- The A320 crash at Mulhouse-Habsheim. To this day the pilot insists it was the aircraft's fault and not his. Ok, so that a/c commanded the engines to go to idle but he knew that was wrong (he should have because he knew how jet engines behave) instead he probably thought to let the a/c do its thing. But it didn't. He in fact should have done what the BA pilot did at White Waltham (see previous post) and that was to 'fly' the a/c down. In that case the engines were never below 80-90%n2.

The BA 777 accident at Heathrow. They got lucky! In that had the engines spooled down just a few seconds earlier they would have come down in a built up area with substantial loss of life. Of course this was a case of excellent airmanship as the engines spooling down was a systems failre due to fuel icing.

And then probably the worst and saddest recent example of lack of airmanship is AF447. None of the crew knew how to react to a basic loss of IAS reading. They totally ignored the stall warnings and worst of all the Captain deliberately chose to leave his post when he knew the a/c was about to fly through very severe weather. He therefore was unable to monitor less experienced crew. In fact had the crew done nothing at all the a/c would have most likely flown itself out of trouble.

Nowadays aircrew are so used to letting the a/c fly itself day in and day out with nothing untoward happening that when something 'unexpected' does occur they are totally out of the loop. At all times the pilot should be flying the a/c in his mind planning ahead for something that may go wrong. On approach thinking about what constitutes a good approach and landing and what does not. And as soon as the a/c systems deviate from his mental picture it's time for manual override!

Modern aircraft with fadec controlled engines do not have problems spooling up from idle. The fadec controls idle speed at an appropriate level for go around spool up within the required time and also for bleed requirements.

Modern aircraft with fadec controlled engines do not have problems spooling up from idle. The fadec controls idle speed at an appropriate level for go around spool up within the required time and also for bleed requirements.

 

Ok,now this is getting worrying, and you say you are a commercial pilot? FADEC or no FADEC, flight idle or ground idle all commercial aircraft have problems spooling up from idle. A CFM56 can take from 5 to 8 seconds to spool from approach/ground idle.   In the case of Asiana it took 8 seconds!

 

This is why stable approach criteria requires a Go-around if the engines are idle,does not  matter if you are on path and flying the correct Vref.  

 

Perhaps I am reading you wrong, but it really sounds like 'The EEC/FADEC takes care of everything' No need to worry.

 

The report clearly shows how slow the 777's engines behaved with the EEC magic in the background

 

 16 seconds prior to impact. At 125 feet and 112 KIAS the thrust levers were advanced and the engines began to spool up 8 seconds prior to impact, the aircraft reached a minimum speed of 103 KIAS 3 seconds prior to impact, the engines were accelerating through 50% engine power at that point

 

Based on the AFDS and autothrottle modes recorded on the FDR data, it is apparent that the thrust levers were manually pushed to full power, rather than by pushing the Take Off Go-around (TOGA) button. The engines began spooling-up immediately, but the normal engine spool-up time from approach idle to full power is seven to eight seconds.

 

Can I ask what you fly? 

Ok,now this is getting worrying, and you say you are a commercial pilot? FADEC or no FADEC, flight idle or ground idle all commercial aircraft have problems spooling up from idle. A CFM56 can take from 5 to 8 seconds to spool from approach/ground idle. In the case of Asiana it took 8 seconds!

 

This is why stable approach criteria requires a Go-around if the engines are idle,does not matter if you are on path and flying the correct Vref.

 

Perhaps I am reading you wrong, but it really sounds like 'The EEC/FADEC takes care of everything' No need to worry.

 

The report clearly shows how slow the 777's engines behaved with the EEC magic in the background

 

 

 

16 seconds prior to impact. At 125 feet and 112 KIAS the thrust levers were advanced and the engines began to spool up 8 seconds prior to impact, the aircraft reached a minimum speed of 103 KIAS 3 seconds prior to impact, the engines were accelerating through 50% engine power at that point

Based on the AFDS and autothrottle modes recorded on the FDR data, it is apparent that the thrust levers were manually pushed to full power, rather than by pushing the Take Off Go-around (TOGA) button. The engines began spooling-up immediately, but the normal engine spool-up time from approach idle to full power is seven to eight seconds.

 

Can I ask what you fly?

Ok, now I know for sure you have never flown a jet, even though you seem to have a lot to say about how to fly an Airbus. Of course you don't want to pull them back to too far when you're dealing with gusts, but if you are getting configured, decelerating, or descending, there is nothing against pulling them back to idle because that is what you need to do. The engines snd their controls now are not like the Conways. Those VC-10 engines probably took at least five minutes to spool up from idle, which is why you kept the levers standing up all the time. I believe 8 seconds is the requirement for spool up from idle to GA thrust in modern designs.

 

I fly the E190.

Ok, now I know for sure you have never flown a jet, even though you seem to have a lot to say about how to fly an Airbus. Of course you don't want to pull them back to too far when you're dealing with gusts, but if you are getting configured, decelerating, or descending, there is nothing against pulling them back to idle because that is what you need to do. The engines snd their controls now are not like the Conways. I believe 8 seconds is the requirement for spool up from idle to GA thrust.

 

I fly the E190.

 

Excuse me but you are the only person here claiming to fly commercially, although a lot of your comments seem to suggest otherwise. I have never claimed to be a commercial pilot in any forum so not quite sure where you got that from?.  My knowledge of Airbus comes from previous work on fixed based simulators and access to the full motion A3xx sim's at the GAA through MENA Aerospace.   I learnt from a young age that it is better to earn enough money to own your own light aircraft than to get paid to fly for someone else.

 

Back on topic, yes it is 8 seconds. You seem to be missing Volibarista's and my point completely.  Modern aircraft 'DO' have issues with spooling up from idle, hense why we have seen so many incidents in the past due to late go-arounds.

 

More importantly these slow spool times and the current certification requirements  are the exact reason why all SOPS require engines spooled at 1000ft at the very latest  IMC or VMC,  and some will allow  500ft VMC.

 

That is no great revelation, your company must operate exactly the same.  And just to be 100% clear, no one here is talking about being idle 'prior' to 1000ft

From memory (don't have the 777 FCOM with me right now) the EEC commands an "approach idle" which is higher than normal (or minimum) idle when Flaps 25 is commanded. This is by design, in order to reduce the time it takes for the engine to reach GA thrust.

 

Evidently this doesn't prevent situations like this from happening, then again it's not the EEC at fault here but the poor energy management by the crew. Had they been flying a stabilized approach using flaps 25 or 30 and the engines would have reached GA thrust with no problems. KSFO 28L/R are both CAT II with DH near the 100ft range, so the engine spool up time issue bit them in their rear due to poor airmanship.

I reckon this is one of the classical case of a Swiss cheese model. I watched the animation they had several opportunities of saving themselves.

 

First of all, if I had to do a non precision approach, I would make use of VNAV and try to stay in VNAV mode throughout the descent from cruise because just like DES mode of the airbus it reduces a huge amount of workload. However VNAV for some reason wasn't used during the approach phase. There's a clear mismanagement and misunderstanding of the airplanes auto flight system. And the decision of going around should have come at 1000' or even earlier.

 

Second, it's a known fact that almost all ATC in the US will keep you high and fast. Going into ORD, I was told 210kts till 10nm and 170kts till 4nm while descending on a Glide Slope on a 77ER with close to max landing weight considering most Asian airports have speed control to 180kts till 8nm and 150-160kts till 4nm, this is quite fast for a 777 which is couple with an increasing head wind as the airplane descend would make it very tough to lower the flaps. Also there's little point to try to achieve continuous desend going to busy airports in the US, just get down ASAP, it reduces workload. I agree ATC could have done a better job, but I wouldn't blame them entirely, because its always the pilots job to ensure safety, ATC had little experience flying a big jet, they would think they are doing the right thing until you speak up. While I would try to accommodate ATC as much as possible as an attempt to make their job slightly easier, I believed every pilot have got a bottom line in their mind, and if in doubt, just reply "standby" to ATC, have a quick discussion wiith the other guy, then make a decision.

 

Third, manual handling skill. One should not sit in the cockpit if he/she can't fly a visual appr with the rwy in sight. The 777 is just an airplane. The industry had already realized the fast decaying of pilot manual handling skills due to huge amount of long haul flyings involved. Also some airlines actually put in policy to restrict pilots from hand flying. Fortunately the Airlines I work for, we have introduced a lot handling simulator sections and people on the line actually are encouraged to fly the airplane manually as much as they want if the conditions fits.

 

The other thing is, when the PF wasn't following the FD, had he switched them off at any point, he would have had SPD mode back and he would be fine. But anyway, I would blame training system. And I wonder what type of simulator exercisesdid the PF gone throught during his type conversion course. And I wonder how he passed his initiate Profficiency check which should include a non precision approach with a visual segment into a non ILS rwy for a landing with raw data and probably should had been done with restricted visbility and ENG inop.

 

It's very easy to blame pilots, I reckon it's the training system Asiana had which failed to train their crew with proper knowledge and skill to the fly the airplane.

Modern aircraft with fadec controlled engines do not have problems spooling up from idle. The fadec controls idle speed at an appropriate level for go around spool up within the required time and also for bleed requirements.

That's not the point!!!!!

Whether an a/c has FADEC or not it still takes time to spool up from idle. And, at low altitude there is not enough space below you. You simply do not let the engines get to idle at low altitude. You are reducing your margin of safety to almost nil! Fadec or no!

 Those VC-10 engines probably took at least five minutes to spool up from idle, which is why you kept the levers standing up all the time. I believe 8 seconds is the requirement for spool up from idle to GA thrust in modern designs.

 

I fly the E190.

Wrong! Very wrong! When the engines are ground tested they must pass the slam test before being certified to fly with the a/c. That is 56%n2 to 105%n2 in 6 seconds. You seem to continously miss the point.

 

It doesn't matter whether a pilot or the computer systems are monitoring the engines. You don't let them get anywhere near to idle on a low approach. You seem to profess total confidence in your a/c systems to the point of not being concerned. That says a lot about your airmanship!

Excuse me but you are the only person here claiming to fly commercially, although a lot of your comments seem to suggest otherwise. I have never claimed to be a commercial pilot in any forum so not quite sure where you got that from?. My knowledge of Airbus comes from previous work on fixed based simulators and access to the full motion A3xx sim's at the GAA through MENA Aerospace. I learnt from a young age that it is better to earn enough money to own your own light aircraft than to get paid to fly for someone else.

 

Back on topic, yes it is 8 seconds. You seem to be missing Volibarista's and my point completely. Modern aircraft 'DO' have issues with spooling up from idle, hense why we have seen so many incidents in the past due to late go-arounds.

 

More importantly these slow spool times and the current certification requirements are the exact reason why all SOPS require engines spooled at 1000ft at the very latest IMC or VMC, and some will allow 500ft VMC.

 

That is no great revelation, your company must operate exactly the same. And just to be 100% clear, no one here is talking about being idle 'prior' to 1000ft

Then quit asking me 'what do you fly?' 'where do you fly?' 'who do you fly for?' if I have nothing to say that makes sense to you. Your attitude is getting a little old.

 

If you and vololiberista had a point with the comments about engine spool up, it was completely irrelevant to this accident. Because, one, the pilots stalled the plane. They were going to crash even if the engines spooled up instantaneously. Second, the pilots did not intend to leave the engines at idle. They had completely abdicated the thrust management to the autothrottles. They probably thought the engines were already spooling up when they finally got on the vasi. Third, those pilots' minds were still stuck way back up the final approach with trying to achieve what they were supposed to have accomplish prior to 1000'; that is getting on slope and on speed. You may not be talking about being above 1000', but I am because I am talking about what these pilots had in their heads.

 

My only point to vololiberista was that modern engines are not flown like Conways. The fadecs will essentially stand the levers up for you as necessary to ensure the required response by setting the idle speed high enough. Depending on conditions, you can see idle speeds from 20% to 55% with the thrust levers all the way retarded in the planes I've flown. We do rejectef landings in training, with a go around being initiated frpm the flare just prior to touchdown. Never had a problem with the engines not spooling up to keep us from hitting the truck the sim instructor always likes to place on the runway.

Wrong! Very wrong! When the engines are ground tested they must pass the slam test before being certified to fly with the a/c. That is 56%n2 to 105%n2 in 6 seconds. You seem to continously miss the point.

 

It doesn't matter whether a pilot or the computer systems are monitoring the engines. You don't let them get anywhere near to idle on a low approach. You seem to profess total confidence in your a/c systems to the point of not being concerned. That says a lot about your airmanship!

Quit yelling. And quit the personal attacks. You and flex1978 both.

 

Those pilots that crashed didn't have any idea whether their engines were spooled up or idle until they were stalling. Engine handling wasn't the problem here. Abdicating engine handling to an autothrottle that wasn't there anymore was the problem.

 

Of course you avoid pulling the engines back to idle once you're stabilized on final. Their problem was that they never were stabilized and they had no idea what their engine was doing. Your point about not going to idle is fine and i'm sure everybody agrees, but is irrelevant here since those guys were not even trying to control the engines.