Moderators: richierich, ua900, PanAm_DC10, hOMSaR

 
BA777ER236
Posts: 170
Joined: Wed Sep 27, 2006 1:18 am

RE: A32x Corrupt Software

Mon Jan 29, 2007 5:42 pm

Quoting Zeke (Reply 42):
The ECL is a step ahead of ECAM, and so it should be

I'm sorry Zeke, I'm not trying to make a negative comparison between the Airbus and the 777. I flew the Airbus for several years before I transferred to the 777, and I thoroughly enjoyed flying the 319/320.

Quoting 777236ER (Reply 43):
Why not the crossfeed system?

Yes, why not, but why also. The system is very simple, the use is simple, and there are very few instances of mistakes using it as pilots take it's use very seriously. Although the 'Air Transat' incident nearly ended in disaster, it didn't, and procedures have now been modified so that it shouldn't happen again. Your intent is an unnecessary expense to develop a complex automated system which would recognise every possible leak scenario etc. To undertake this, the manufacturers would want to have a demonstrable improvement in safety margins which I just don't see is there.

Quoting 777236ER (Reply 43):
On a future aircraft, the fuel system could be completely transparent and it would improve flight safety.

I don't agree! Several of us have quoted incidents on this thread where 'automatic' systems have failed, but you are determined to take everything out of the loop for the pilot. The net result will be that the crews will not perceive an insidious 'automatic' failure, and end up in a difficult position. Your assertion that automation improves flight safety is therefore highly dubious, and merely an opinion that needs to be backed up by statistics.

Quoting 777236ER (Reply 50):
their failure to select it to ALTN immediately

There was no 'failure' here. The crew followed the ECAM actions as they were trained to do and as they should have done. When they got to that part of the ECAM, they selected to ALTN, and the majority of the systems came back. I don't think there is an argument about the way the BA crew handled the systems side of the emergency, only an argument over their subsequent actions!

Quoting 777236ER (Reply 50):
The level of automation in the 777 would have prevented that A319 from flying around without radios. And there's still scope for improvement.

The two systems are different in many ways, and do not draw direct comparison. Electrical systems are highly complex, and tend to produce very 'interesting' scenarios at times. One of the reasons that the AC ESS push button is there, is to guard agains't a suspected a short in the AC ESS bus, and NOT to connect it to a good power supply only to short that!

Quoting Pihero (Reply 49):
You really don't see our point, because you are too involved in your demonstration.

Absolutely, and good to see you here Pihero.

What this boils down to, is that engineers (not all, I know!) are determined to design the pilot out of the flight deck, and pilots are determined to keep them in and to have more involvement. The day that fully automated commercial aeroplanes start to fly passengers is the day when I will stop flying permanently (as a pax!)

Cheers
 Smile
Flying would be easy if it wasn't for the ground
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Mon Jan 29, 2007 8:57 pm

Quoting BA777ER236 (Reply 51):

Yes, why not, but why also. The system is very simple, the use is simple, and there are very few instances of mistakes using it as pilots take it's use very seriously. Although the 'Air Transat' incident nearly ended in disaster, it didn't, and procedures have now been modified so that it shouldn't happen again.

It's by the grace of God that the 306 people on board that A330 lived. Consider five of the findings of the report:

8. The flight crew did not detect that a fuel problem existed until the Fuel ADV advisory was displayed and the fuel imbalance was noted on the Fuel ECAM page.
9. The crew did not correctly evaluate the situation before taking action.
10. The flight crew did not recognize that a fuel leak situation existed and carried out the fuel imbalance procedure from memory, which resulted in the fuel from the left tanks being fed to the leak in the right engine.
11. Conducting the FUEL IMBALANCE procedure by memory negated the defence of the Caution note in the FUEL IMBALANCE checklist that may have caused the crew to consider timely actioning of the FUEL LEAK procedure.
12. Although there were a number of other indications that a significant fuel loss was occurring, the crew did not conclude that a fuel leak situation existed - not actioning the FUEL LEAK procedure was the key factor that led to the fuel exhaustion.


The problem wasn't lack of procedures, or even lack of information. The crew took action before they looked at all the possibilities, and they conducted the actions from memory - this in an era of a strict adherance to the checklists and QRHs and CRM. What's to stop another crew from carrying out memory checklists when it's not appropriate over the Atlantic today? Nothing. The Air Transat crew knew they should have looked at the checklists and didn't.

Had the A330 crashed it would have been a huge event. Imagine a current-generation twin aircraft crashing into the Atlantic whilst on an ETOPS flight. If it had crashed I think you'd have seen major revisions to the fuel system software. I don't think we should let the fact it was a near-disaster stop us from progressing.

A fuel control computer would have recognised much earlier that there was a fuel leak. It wouldn't have initiated the crossfeed. It could then have told the crew the situation exactly, telling them how much time they had til a fuel emergency.

Can you name any situation where a flight crew would need manual control of a crossfeed valve?

Quoting BA777ER236 (Reply 51):
Several of us have quoted incidents on this thread where 'automatic' systems have failed, but you are determined to take everything out of the loop for the pilot.

But you fly an aircraft that's even more automated than the A320, has no physical link from the flight controls to the primary or secondary control surfaces and has taken the pilot out of the loop of most systems! You're trusting your life to automated systems as it is. My suggestions aren't as dramatic as fly by wire, FADEC or sidesticks.

Quoting BA777ER236 (Reply 51):

There was no 'failure' here. The crew followed the ECAM actions as they were trained to do and as they should have done. When they got to that part of the ECAM, they selected to ALTN, and the majority of the systems came back. I don't think there is an argument about the way the BA crew handled the systems side of the emergency, only an argument over their subsequent actions!

By no means am I critisising the BA crew (though maybe the easyJet ones  Wink, time will tell if they actually pressed the AC ESS p/b!). My point is, having an A319 flying above the UK with no power to the flight deck for two minutes is Bad with a capital B. An electrical system computer would have detailed knowledge of the state of the system and realise that the AC ESS bus was unpowered, due to the AC1 bus being unpowered, but the AC2 bus is powered to switching the AC ESS supply would sort the problem out. There could potentially be no percieved interuption in power to the crew. That's surely better than an aircraft losing the essential AC busbar and relying on the crew to carry out the right checklist (or not mess up a memory checklist like the Air Transat crew did) and restore power.

In the easyJet case, an electrical system computer could restore the radios - having an A319 flying around with no RTF is surely not acceptable. In that case the crew had good training, good CRM and good checklists, but for some reason power wasn't restored (not to preempt the AAIB report, but I presume that's because they didn't press the AC ESS p/b because it was unlit and so couldn't determine its status).

I don't see how you can really argue with these last points - your aircraft essentially does this already.

Quoting BA777ER236 (Reply 51):

What this boils down to, is that engineers (not all, I know!) are determined to design the pilot out of the flight deck, and pilots are determined to keep them in and to have more involvement.

It is true that pilots want to stay in the flight deck and engineers want to get rid of them! But that's not what I'm advocating. I'm not talking about removing two pilots from the flight deck. Like it or not, the automation of systems over the past 30 years has brought improvements in flight safety.

You can say that I'm too wrapped up in my examples, but they're still specific examples of where greater automation would have improved flight safety. The crossfeed switches aren't required, AC ESS p/bs aren't required.
Your bone's got a little machine
 
BA777ER236
Posts: 170
Joined: Wed Sep 27, 2006 1:18 am

RE: A32x Corrupt Software

Mon Jan 29, 2007 10:58 pm

Quoting 777236ER (Reply 52):
You can say that I'm too wrapped up in my examples, but they're still specific examples of where greater automation would have improved flight safety.

As you seem to be focused on automating the Fuel system to take out the x-feed based on this non-fatal incident, consider this:- The 757/767 and 777 fuel systems are very similar in their operation. I think you'll agree that the entire fleet of these aeroplanes has flown, since 1982, millions of hours, and hundreds of thousands of sectors. I can't recall a single incident where the crews mishandling of the Fuel panel lead to an accident on these fleets - can you?
Given that sort of record, why change it?

Cheers
 Smile
Flying would be easy if it wasn't for the ground
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Tue Jan 30, 2007 2:01 am

Quoting BA777ER236 (Reply 53):
As you seem to be focused on automating the Fuel system to take out the x-feed based on this non-fatal incident, consider this:- The 757/767 and 777 fuel systems are very similar in their operation. I think you'll agree that the entire fleet of these aeroplanes has flown, since 1982, millions of hours, and hundreds of thousands of sectors. I can't recall a single incident where the crews mishandling of the Fuel panel lead to an accident on these fleets - can you?
Given that sort of record, why change it?

Prior to 2001 how many millions of flight hours were flown by aircraft with crossfeed valves? Who'd have thought that a flight crew would crossfeed fuel from a full tank into an emptying tank that had a leak in it?

The industry needs to learn from its mistakes. There's nothing stopping any aircraft crew from making the same mistakes, save knowledge of what happened to the A330. What about in 5, 10, 20 years when crew have forgotten about the Air Transat situation?

It comes down to whether two stressed crew are better at dealing with a developing emergency situation, with limited information, limited knowledge and limited control, or whether a flight computer programmed by dozens of engineers on the ground is better. The trend has been towards the latter, and safety has improved because of it.
Your bone's got a little machine
 
Pihero
Posts: 4318
Joined: Mon Jan 31, 2005 5:11 am

RE: A32x Corrupt Software

Tue Jan 30, 2007 2:40 am

Quoting 777236ER (Reply 52):
My suggestions aren't as dramatic as fly by wire, FADEC or sidesticks

Is it not ? To me, FBW is just an extension of the flight controls that pre dated it : a pilot's order--> an aircraft response, exactly the same over and over...then someone added the guard rails...still simple programming :basically a third degree system solver, a Zero reader plus a simple loop.

What you are advocating is an other order of magnitude and we're close -if not completely inside - A.I.

Funny you did not pick up my experience of the electrical arcing on my windscreen. We are exactly where you want things (away from that idiot Pihero and his big clumsy fingers). Problem was the very intelligent engineer who designed the window heat system and put an automatic power on linked to the air/ground sensor forgot to imagine that scenario.
I don't know whether you had been facing an impending windscreen shattering because you cannot remove the power from that window and you sit there watching the different glass layers bubbling away...
Did I tell you that in the 6 minutes it took me to bring that airplane back down , we'd already lost the outer layer ? And did I tell you that I still have the shivers imagining the same problem at 39,000 ft (I would have switched all power off and done a bus search).

Taking that experience in mind, I cannot but find arrogant the engineer's attitude. They've thought of everything, indeed !
Remember when they sold the 320 in the beginning ? "It's the wonder that will prevent pilots from making silly mistakes" ! Result : two very silly - and fatal - mistakes from people who were sold on the advert , one in Habsheim, one in Bangalore... until some cool heads started to calm things down and put the new technology into some sort of perspective.

In my very HUMBLE opinion, that's what is needed.
Contrail designer
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Tue Jan 30, 2007 3:02 am

Quoting Pihero (Reply 55):
Is it not ? To me, FBW is just an extension of the flight controls that pre dated it : a pilot's order--> an aircraft response, exactly the same over and over...then someone added the guard rails...still simple programming :basically a third degree system solver, a Zero reader plus a simple loop.

What you are advocating is an other order of magnitude and we're close -if not completely inside - A.I.

Not at all. Consider the A320, there's a static inverter which will supply DC power from Battery 1 to the AC ESS bus. The crew have absolutely no control over this. It's a system which could save lives and the crew can do nothing with it except maybe pull a c/b to deactivate it. It will start automatically and stop automatically.

The AC ESS p/b could be removed and replaced with a system that's along the same lines. It's not a feat of computer engineering - the autoflight computers are still much more sophisticated.

Quoting Pihero (Reply 55):
Funny you did not pick up my experience of the electrical arcing on my windscreen. We are exactly where you want things (away from that idiot Pihero and his big clumsy fingers). Problem was the very intelligent engineer who designed the window heat system and put an automatic power on linked to the air/ground sensor forgot to imagine that scenario.
I don't know whether you had been facing an impending windscreen shattering because you cannot remove the power from that window and you sit there watching the different glass layers bubbling away...
Did I tell you that in the 6 minutes it took me to bring that airplane back down , we'd already lost the outer layer ? And did I tell you that I still have the shivers imagining the same problem at 39,000 ft (I would have switched all power off and done a bus search).

I did pick up on your experience and suggested that the problem was the automatic systems don't tend to have automatic ways of dealing with failures. In the arcing example, it would be a simple solution to put a current monitor in the system that could detect arcing. You seem to be complaining about the current level of automation, let alone the future.

Quoting Pihero (Reply 55):
Remember when they sold the 320 in the beginning ? "It's the wonder that will prevent pilots from making silly mistakes" ! Result : two very silly - and fatal - mistakes from people who were sold on the advert , one in Habsheim, one in Bangalore... until some cool heads started to calm things down and put the new technology into some sort of perspective.

The changes in the A320 flight computer software weren't radical after either of those crashes. The problems weren't as a consequence of automation, but rather poor training. With more systems automation it becomes less likely that crew can make fatal mistakes.

I don't know why flight crews take any sort of cockpit, flight controls or control philosophy argument so personally. The truth is that crew error is still one of the leading causes of aircraft crashes. Flying isn't an ego boost for pilots, it's a job where safety is paramount. There's no need for pilots to take as a personal insult the mere suggestion that the level of automation in aircraft isn't at its peak.
Your bone's got a little machine
 
BA777ER236
Posts: 170
Joined: Wed Sep 27, 2006 1:18 am

RE: A32x Corrupt Software

Tue Jan 30, 2007 5:13 am

Quoting 777236ER (Reply 56):
Flying isn't an ego boost for pilots, it's a job where safety is paramount. There's no need for pilots to take as a personal insult the mere suggestion that the level of automation in aircraft isn't at its peak.

Neither is it an ego boost for engineers who would needlessly over complicate everything. You just don't seem to understand that we DON'T need a multi logic/scenario fuel system which will just be another system that can fail! You keep on banging on about safety being improved, but you can't just base that on one isolated incident! The lessons learnt from that particular incident will survive because they are written into ECAM/ECL and QRH's and they are discussed in CRM courses.

Professional pilots like Pihero, Zeke and myself are very well aware that we don't do this job for our egos, and we devote many hours to technical and CRM study, and I used to teach CRM as part of my training duties. Frankly, I find it insulting that you consider my attitude to this as being related to an ego trip!

Instead of designing further 'under the skin' transparency, engineers should be concentrating on designing better involvement of pilots in the operating scenario instead of proposing an unneeded complication with some vague notion of an 'improvement' in flight safety.

Cheers
 Smile
Flying would be easy if it wasn't for the ground
 
prebennorholm
Posts: 7093
Joined: Tue Mar 21, 2000 6:25 am

RE: A32x Corrupt Software

Tue Jan 30, 2007 10:33 am

This happens all the time. Two weeks ago a colleague of mine boarded a B744 in Amsterdam only to board another 744 six hours later and arrive six hours late in Houston, TX. Reason: Broken computer.

He was not compensated. And neither do I assume that Boeing compensated KLM. (Who knows, maybe KLM just fouled up a software update on that particular plane?)

Quoting Rpaillard (Reply 4):
Well, my VW dealer never compensate me for various electronics problem in my car. They fix it for free.

Lucky you! My VW dealer fixed my 1998 VW right hand side side-airbag after 7 years and 340,000 km (210,000 miles). After four attempts at a total cost for me around $1k. It helped when I told him that I was running out of black tape to cover the airbag fault indicator light.
Always keep your number of landings equal to your number of take-offs
 
CHRISBA777ER
Posts: 3715
Joined: Thu Mar 01, 2001 12:12 pm

RE: A32x Corrupt Software

Tue Jan 30, 2007 8:18 pm

777236ER - tell me, what do you fly?
What do you mean you dont have any bourbon? Do you know how far it is to Houston? What kind of airline is this???
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Tue Jan 30, 2007 9:49 pm

Quoting CHRISBA777ER (Reply 59):
777236ER - tell me, what do you fly?

I never understood the attitude and complete arrogance of pilots, who feel that they are top of the food chain in the aerospace industry. According to http://www.planecrashinfo.com/cause.htm (and feel free to find another reference), pilot error is by far the biggest cause of aircraft crashes. Unless flight crews are willing to openly work with engineers, then an improvement in flight safety and your safety won't come.

You still haven't told me one case where the pilot would do a better job using the crossfeed p/b than a computer would controlling the valve.

Quoting BA777ER236 (Reply 57):
You just don't seem to understand that we DON'T need a multi logic/scenario fuel system which will just be another system that can fai

You could have made exactly the argument 10, 20, 30, 40, 50 years ago about any of the systems on the 777 - yet you still think the 777 is the pinacle. Tell me, do you disapprove of Boeing's standard HUD in the 787? Or increased LCD space? Or increased automation? Or deletion of bleed air systems?

Quoting BA777ER236 (Reply 57):
You keep on banging on about safety being improved, but you can't just base that on one isolated incident!

Why not? That 'one isolated incident' would have cost the lives of 300 people, but by the grace of God. The only thing preventing it happening again is an airwareness of crews of that incident and crew training. The awareness won't last forever, and as incident after incident shows, crew can ignore training. Computers can't ignore programming.

Quoting BA777ER236 (Reply 57):
The lessons learnt from that particular incident will survive because they are written into ECAM/ECL and QRH's and they are discussed in CRM courses.

In the Air Transat case, the QRH was clear that the crew should consider the possibility of a fuel leak - they didn't. One competant crew flying a current-generation aircraft with automation and ECAM and CRM and good training didn't pick up on what was going wrong - it will happen again.

Quoting BA777ER236 (Reply 57):
Frankly, I find it insulting that you consider my attitude to this as being related to an ego trip!

So why are you so averse to change? Deleting the crossfeed p/b will prevent crews from ever feeding fuel into a tank with a leak in it. Why is that such a bad thing? Crossfeed logic isn't complicated - the program required to control it would be much less complicated than the FBW programmes or autoflight programmes. The level of control lost with the deletion of that p/b is much less than the level of control lost between the 727 and the 737NG or the 747-100 and the 747-400, say.

Quoting BA777ER236 (Reply 57):

Instead of designing further 'under the skin' transparency, engineers should be concentrating on designing better involvement of pilots in the operating scenario

So you propose a move back towards 727-style flight decks, with almost no automation where the crew can control every from pitot heat and fuel heat to the yaw damper? The pilots had almost complete involvement in aircraft systems for decades. With increased automation came increased safety, you can't deny that.
Your bone's got a little machine
 
User avatar
zeke
Posts: 15265
Joined: Thu Dec 14, 2006 1:42 pm

RE: A32x Corrupt Software

Tue Jan 30, 2007 11:15 pm

Quoting 777236ER (Reply 56):
The problems weren't as a consequence of automation, but rather poor training.

The two air france training captains flying the aircraft at Habsheim we probably some of the most experienced and trained people on the 320. That accident would have happened in any jet put in the same situation, low altitude, low airspeed, high angle of attack, and idle thrust, with obstacles looming at a height higher than they were flying.

Quoting 777236ER (Reply 56):
There's no need for pilots to take as a personal insult the mere suggestion that the level of automation in aircraft isn't at its peak.

I dont feel like I get overloaded in the the sim when they throw numerous problems at me, it just a matter of time management, so personally I dont feel like I need more automation, but I will gladly take it if someone wants to give it to me. I can happily now do a one engine out ILS with autothrust and autopilot in, so to me this it what I really need to fly the aircraft.

Whilst not automation per say, I think some more effort could be made in the area of ECAM to simplify things more for crews, this could include the QRH procedure text, and automatic calculation of landing distance and speed increments in failure situations.

Systems themselves can only be as automated to the extent that the back end hardware allows, with say the 320 electrical system we are now looking at a system design which is 20 years old. Better technology is about now which allows for different levels of redundancy and different designs, however I think for commonality they system design will have to remain as is.

My two biggest concerns with a computerised aircraft is not FBW, electronic checklists, electronic screens, it is a fire or smoke in the avionics bay and loss of situational awareness. I need the ability from the cockpit to isolate different parts of electrical system long enough to get an aircraft on the ground and get my passengers out, and I would love to have a terrain chart like an ONC on the ND with my track and navaids overlaid. EGPWS is an improvement, however we have only just recently had another carrier just miss hitting hill on departure here at HKG as they turned too early not reading the chart, and by luck, not design, managing to go in a saddle between two hills clearing the saddle by a few hundred feet with the peaks higher on either side.

This brings me to my last point, the most dangerous time for us is when we are on the ground, just leaving or just about to land. I think the reason for many of the accidents in these phases goes back not to solely to the automation inside the cockpit, but also the the antiquated signage, markings, approach and departure procedures. More in cockpit situational awareness needs to be available on the ground with moving maps of the actual position on the airport, and clearances need to be issued digitally to overlay that on that map. In my view the main problem in this phase are lack of standard communication and frequency congestion.

On departure we need to be able to see safe corridors where we can fly, so some person in a radar room somewhere does not vector us into a hill. EGPWS just not work that well for turning flight.

I still do not know why the performance calculations are not made onboard with the real aircraft mass and cg, fairly simple maths.

For arrival we need better approaches, for jets still to be doing night circling approaches is ridiculous, everything should be runway aligned with a glidepath of some type. All approach charts should be overlaid with actual terrain on the ND.

Quoting 777236ER (Reply 60):
You still haven't told me one case where the pilot would do a better job using the crossfeed p/b than a computer would controlling the valve.

Non standard problems, like a problem with the pump, MEL etc.

Quoting 777236ER (Reply 60):
Tell me, do you disapprove of Boeing's standard HUD in the 787?

To me its more of a gimmick.

Quoting 777236ER (Reply 60):
Or increased LCD space?

Have not seen an accident caused by a small screen.

Quoting 777236ER (Reply 60):
Or increased automation?

Automation is about the same as a 777, and to some extent less than the 380.

Quoting 777236ER (Reply 60):
Or deletion of bleed air systems?

Still has bleed, just not for cabin pressure. The systems are similar, just the generation of pressure is different, the pressurisation controllers are basically the same as the 380 ones.

Quoting 777236ER (Reply 60):
One competant crew flying a current-generation aircraft with automation and ECAM and CRM and good training didn't pick up on what was going wrong - it will happen again.

 ziplip 

Quoting 777236ER (Reply 60):
The pilots had almost complete involvement in aircraft systems for decades. With increased automation came increased safety, you can't deny that.

I think the statistics show getting the pilots have control of aircraft systems and getting rid of flight engineers has improved safety.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
User avatar
Starlionblue
Posts: 20175
Joined: Fri Feb 27, 2004 9:54 pm

RE: A32x Corrupt Software

Wed Jan 31, 2007 1:45 am

Quoting Zeke (Reply 61):
This brings me to my last point, the most dangerous time for us is when we are on the ground, just leaving or just about to land. I think the reason for many of the accidents in these phases goes back not to solely to the automation inside the cockpit, but also the the antiquated signage, markings, approach and departure procedures. More in cockpit situational awareness needs to be available on the ground with moving maps of the actual position on the airport, and clearances need to be issued digitally to overlay that on that map. In my view the main problem in this phase are lack of standard communication and frequency congestion.

Interesting perspective. Would you say that the planes are getting every more complex while ground control/ATC is not keeping up technology/SOP wise?
"There are no stupid questions, but there are a lot of inquisitive idiots." - John Ringo
 
Pihero
Posts: 4318
Joined: Mon Jan 31, 2005 5:11 am

RE: A32x Corrupt Software

Wed Jan 31, 2007 2:27 am

Quoting 777236ER (Reply 60):
pilot error

here we are !
And of course that final verdict cannot be challenged, can it ?
Instead of just poring into some sensationalist site, why don't you go to the Bielefeld University research on automation / human factors, or the RISK project... to name just two.

Generally one changes one's mind after an afternoon with these people.
For instance an error is not an isolated event in the course of some well oiled, smooth process...It has a set of reasons and very often stems from a faulty, inaccurate, difficult to interpret, or even missing original information.
We can argue at length about ergonomics and man-machine interface, history shows that instrument designers, cockpit ergonomists share a big deal of fatal errors...to this very day. Witness the three-pointed altimeter....to the TCAS "monitor vertical speed" resolution command...the automatic selection of one waypoint in a long list of possibles on an ND...
And now the same people are saying get rid of those guys, we will be a lot safer ?
I have talked at length about the electrical arcing in a windscreen...The system is exactly as you want it :FULLY AUTOMATIC :you takeoff and it goes on, you land and it goes off, am I not correct ?
The designer forgot that there could be some situations where the power to these windows SHOULD be removed and I could have been, as a result of that overlook another statistics.

In the same post you talk about, among other systems, a head-up display.
The fact that you didn't see that : a HUD + enhanced vision system + a simple flight director = a possibility of a manual cat III B landing.

And there is where our respective arrogances differ : We want better tools, you want.... by the way, what do you want ?

Regards
Contrail designer
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Wed Jan 31, 2007 4:35 am

Quoting Zeke (Reply 61):
The two air france training captains flying the aircraft at Habsheim we probably some of the most experienced and trained people on the 320. That accident would have happened in any jet put in the same situation, low altitude, low airspeed, high angle of attack, and idle thrust, with obstacles looming at a height higher than they were flying.

That's my point. Despite the immense training and experience the crew had, they still put the aircraft into an unsafe, low-energy condition.

Quoting Zeke (Reply 61):
My two biggest concerns with a computerised aircraft is not FBW, electronic checklists, electronic screens, it is a fire or smoke in the avionics bay and loss of situational awareness. I need the ability from the cockpit to isolate different parts of electrical system long enough to get an aircraft on the ground and get my passengers out, and I would love to have a terrain chart like an ONC on the ND with my track and navaids overlaid. EGPWS is an improvement, however we have only just recently had another carrier just miss hitting hill on departure here at HKG as they turned too early not reading the chart, and by luck, not design, managing to go in a saddle between two hills clearing the saddle by a few hundred feet with the peaks higher on either side.

This brings me to my last point, the most dangerous time for us is when we are on the ground, just leaving or just about to land. I think the reason for many of the accidents in these phases goes back not to solely to the automation inside the cockpit, but also the the antiquated signage, markings, approach and departure procedures. More in cockpit situational awareness needs to be available on the ground with moving maps of the actual position on the airport, and clearances need to be issued digitally to overlay that on that map. In my view the main problem in this phase are lack of standard communication and frequency congestion.

From my point of view as a systems engineer - and bear that in my, because I'm not a pilot - is that automating systems to the highest possible extent allows the crew to do what they want to do, what they're trained to do and frankly, what they exist to do - that's fly the aircraft.

I would think it's much better for the crew to be heads up, looking out the aircraft or at primary flight instruments and having situational awareness of how the aircraft is actually flying, rather than situational awareness of systems that can be automated. If you reduce the burden of the systems on the crew you free up the resources of the crew to do other things.

Airbus have always done this, and are still doing it with the A380, eg. automatic flap retraction, runway turnoff entry on the MCDU in order to calculate autobraking requirement. It's a lot better for crew to be focussed on flying the aircraft in a difficult, manually flown SID (say) or a difficult V2 climb at a difficult airport after an engine failure, rather than worrying about F and S speeds etc.

Quoting Zeke (Reply 61):
For arrival we need better approaches, for jets still to be doing night circling approaches is ridiculous, everything should be runway aligned with a glidepath of some type. All approach charts should be overlaid with actual terrain on the ND.

The drive from us at least (though the regulators are more cautious) is to increase the use of LNAV/VNAV in approaches and really try and push for more GLS approaches with LAAS and WAAS. The state of play with regards to approaches is pretty ridiculous, considering the advances made in other areas of navigation and control, and even though it's not particularly relevent to me it's a hot potato within the industry.

Considering the inherant difficulty of approaches, the high traffic volume, the potential for emergencies to come to a head (fuel emergencies etc), the integration of different systems in a very short period of time (flaps, gear, braking) and the historical risks associated with approaches and landing (most dangerous phase of flight), improved systems automation will improve flight safety.

Consider an aircraft making a CAT 1 approach to an airport with the ceiling at 800ft, RVR of 800m. At 400ft a similar situation to the BA and easyJet incidents happen - the crew lose all displays except the lower ECAM/EICAS, standby instruments are unlit, the autopilot disconnects and the flight deck lights go out. Do you go around, an enter cloud with an aircraft that could have significant damage, no function PFD or lights and almost no electrics and no ILS reciever? Or do you land with attitude and airspeed indications coming from a small, unlit - possibly unpowered - standby gauge and the only assurance of braking coming from a small, unlit accumulator gauge?

Isn't it much better to have an electrical system that will automatically reconfigure itself in any given situation, regardless of how simple it would be for the crew to do it manually?

Quoting Zeke (Reply 61):
Non standard problems, like a problem with the pump, MEL etc.

But I still don't see why those events couldn't be incorporated into a computer. The crossfeed system isn't the most complicated, by a long way. There's only a finite number of things that can go wrong with it.

Quoting Zeke (Reply 61):
To me its more of a gimmick.

Well yes, that's probably true, but airlines seem to like them. Regardless it shows the trend to get the pilots to become more like pilots and less like systems engineers. Remember that Airbus are also going down the HUD route too.

Quoting Zeke (Reply 61):
Have not seen an accident caused by a small screen.

True, but you yourself mentioned you'd like increase situational awareness with regards to navigation. Larger screens means better EGPWS information, more chance to have vertical clearance information. And remember that a contributing factor to Swissair 111 was the small size of the standby gauges.

Quoting Zeke (Reply 61):
Still has bleed, just not for cabin pressure. The systems are similar, just the generation of pressure is different, the pressurisation controllers are basically the same as the 380 ones.

Really? I thought there was no bleed from the 787 GEnx and Trent 1000 at all? Isn't deicing being done electrically? The consensus is with us that bleedless systems have almost no performance or weight advantages. In fact, the main advantage is getting rid of a source of spurious engine fire warnings, really. Anyway, the point is that there is still a drive to simplify (and eliminate) systems.

Quoting Zeke (Reply 61):
I think the statistics show getting the pilots have control of aircraft systems and getting rid of flight engineers has improved safety.

Hmm, not sure I agree with that. Certainly flight engineers had a more indepth knowledge of aircraft systems than crews do currently. It's difficult to say the impact the loss of flight engineers had to flight safety because with their elimination came a mass automation of systems.

Quoting Pihero (Reply 63):
I have talked at length about the electrical arcing in a windscreen...The system is exactly as you want it :FULLY AUTOMATIC :you takeoff and it goes on, you land and it goes off, am I not correct ?
The designer forgot that there could be some situations where the power to these windows SHOULD be removed and I could have been, as a result of that overlook another statistics.

In the same post you talk about, among other systems, a head-up display.
The fact that you didn't see that : a HUD + enhanced vision system + a simple flight director = a possibility of a manual cat III B landing.

I've never ever talked about getting rid of pilots. I think firmly that two crew having a significant input flying is the best option. However, there are ways in which controling an aircraft can be better.

Manual CATIII landings are better than automatic ones, hence the inclusion of the HUD. Pilots can spot trends of external problems, they're more engaged with the approach, they keep up their landing competancy. What you don't want is crew performing manual approaches whilst worrying about systems in the aircraft.

In the arcing example, I think it would be better if the system controlling the heat had a current detector and could detect the arcing condition and automatically switch off the system. What's better - arcing occuring at V1 with the crew trying to fly the aircraft and figure out what's going on and correcting the issue, with one or both of the crew going heads down to try and correct the problem, or a system automatically detecting the fault, shutting off power and allowing the crew to fly the airplane?

Automate systems and you give the crew more freedom to actually fly the aircraft. Airbus did this with the A320. They automated systems, simplified systems and crucially did away with controls that selected positions of control surfaces to controls that defined a specific pitch and bank rate. It gave the pilots much more scope to fly the aircraft how they wanted.

My point is that this idea hasn't matured yet.

Edit: is it just me or is the forum really screwed up today?

[Edited 2007-01-30 20:35:28]
Your bone's got a little machine
 
User avatar
zeke
Posts: 15265
Joined: Thu Dec 14, 2006 1:42 pm

RE: A32x Corrupt Software

Wed Jan 31, 2007 2:20 pm

Quoting 777236ER (Reply 63):
Despite the immense training and experience the crew had, they still put the aircraft into an unsafe, low-energy condition.

It was an organisation problem in my view, the aircraft should have never been there in the first place, and no pax should have been onboard either.

If the show was successful, would it solely have been attributed to the pilots ?

Quoting 777236ER (Reply 63):

Isn't it much better to have an electrical system that will automatically reconfigure itself in any given situation, regardless of how simple it would be for the crew to do it manually?

Nope, if the electrical system is at fault, the last thing I want is for it to reconfigure, suggest by all means.

Quoting 777236ER (Reply 63):
There's only a finite number of things that can go wrong with it.

I would disagree, Air Transit brings the point home, that incident was an organisation problem before the aircraft had left it point of departure.

Quoting 777236ER (Reply 63):
Larger screens means better EGPWS information, more chance to have vertical clearance information.

The side profile information in my view is junk, does not help me if I want to turn. The harrier has had moving topo map for years.

Quoting 777236ER (Reply 63):

Really? I thought there was no bleed from the 787 GEnx and Trent 1000 at all?

Engine anti-ice is still via bleed.

Quoting 777236ER (Reply 63):
In fact, the main advantage is getting rid of a source of spurious engine fire warnings, really.

Never had one in my life.

Quoting 777236ER (Reply 63):

Manual CATIII landings are better than automatic ones, hence the inclusion of the HUD.

Hmm no, CAT 3 landings should always be fully automatic, only aircraft that need a HUD to do them manually do not have the ability for runway rollout. 320/330/340 has had the option of HUD for years, not that I have seen any installed, just not needed on an aircraft that is CAT 3B, dash 8 on the other hand, need a HUD to get the approval.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
BA777ER236
Posts: 170
Joined: Wed Sep 27, 2006 1:18 am

RE: A32x Corrupt Software

Wed Jan 31, 2007 7:41 pm

Quoting 777236ER (Reply 63):
I would think it's much better for the crew to be heads up, looking out the aircraft or at primary flight instruments and having situational awareness of how the aircraft is actually flying, rather than situational awareness of systems that can be automated

I don't think that any pilot would disagree with your sentiment here, but situational awareness of systems is also vital whether they are automated or not.

Quoting 777236ER (Reply 63):
But I still don't see why those events couldn't be incorporated into a computer. The crossfeed system isn't the most complicated, by a long way. There's only a finite number of things that can go wrong with it.

Well, I don't have the same confidence in computers (nor their programmers) that you do. You keep asking for examples:- The 777, grand aeroplane that it is, is not perfect. It does suffer occasionally from a problem that we call 'mid Atlantic drop off'. This is where we get a discrepancy between the total indicated fuel and the FMC calculated fuel. Often this is due to one of the many fuel tank sensors not measuring correctly, or the fuel density corrections being slightly off. It often affects one wing tank and not the other, and (almost) always corrects itself before landing. It occasionally triggers the ECL 'Fuel Imbalance' checklist, and whilst we would not ignore a checklist, it is a matter of judgement as to whether you have a fuel leak or whether it is the above problem, and as to how much you restore the balance knowing that the imbalance may be spurious. This is a clear example of a situation where I would not want an automatic crossfeed system. In this case, it would probably needlessly attempt to balance. At best, this may be pointless, and at worst, it could be highly misleading, either masking the problem or erroneously assuming a fuel leak.

Given the above scenario, I would not be happy with an automatic crossfeed system that did not have a manual override. In which case, why have an automatic system at all.


Quoting 777236ER (Reply 60):
One competant crew flying a current-generation aircraft with automation and ECAM and CRM and good training didn't pick up on what was going wrong - it will happen again.

Yes quite Zeke:-  ziplip 

Quoting 777236ER (Reply 59):
Crossfeed logic isn't complicated - the program required to control it would be much less complicated than the FBW programmes or autoflight programmes.

No, I think it is complicated, see my example above.

You seem to think that us pilots are a constant threat to flight safety whereas with careful engineering design and thoughtful ergonomic design we can enhance flight safety from both directions. Please don't design and incorporate systems just for the sake of it though.

Cheers
 Smile
Flying would be easy if it wasn't for the ground
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Wed Jan 31, 2007 9:42 pm

Quoting Zeke (Reply 64):
Nope, if the electrical system is at fault, the last thing I want is for it to reconfigure, suggest by all means.

So you would prefer the situation where screens go dark, lights go out, p/bs are unlit and the restoration of power comes down to the crew pressing the AC ESS p/b - when the recent evidence has shown clear that crew sometimes won't do that?

Quoting Zeke (Reply 64):
I would disagree, Air Transit brings the point home, that incident was an organisation problem before the aircraft had left it point of departure.

That's a bit of a cop out, most crashes can be partly attributed to 'organisational failures' before flight leaving the aircraft or crew in a dangerous position. But regardless of that point, there are only a finite number of failure scenarios with any system on an aircraft. This is only logic.

Quoting Zeke (Reply 64):
Engine anti-ice is still via bleed.

Oh ok - still a pretty radical simplification of the pneumatic systems (even if the total efficiency savings are essentially nil).

Quoting Zeke (Reply 64):
Never had one in my life.

Oh well then, not worth considering is it! The fact is that bleed air leaks cause spurious engine fire warnings, your experience notwithstanding.

Quoting Zeke (Reply 64):

Hmm no, CAT 3 landings should always be fully automatic,

So what point are you even arguing? You seem to think that matters relating to flight should be more automated, but matters relating to sytems should be less automated? Do you think the A320/330/340 are the pinnacle of automation, or the 777, or the A380? Are you sure you're not just arguing the toss because of some imaginary slight you've percieved against Airbus somewhere?

Quoting BA777ER236 (Reply 65):
The 777, grand aeroplane that it is, is not perfect. It does suffer occasionally from a problem that we call 'mid Atlantic drop off'. This is where we get a discrepancy between the total indicated fuel and the FMC calculated fuel. Often this is due to one of the many fuel tank sensors not measuring correctly, or the fuel density corrections being slightly off. It often affects one wing tank and not the other, and (almost) always corrects itself before landing. It occasionally triggers the ECL 'Fuel Imbalance' checklist, and whilst we would not ignore a checklist, it is a matter of judgement as to whether you have a fuel leak or whether it is the above problem, and as to how much you restore the balance knowing that the imbalance may be spurious. This is a clear example of a situation where I would not want an automatic crossfeed system. In this case, it would probably needlessly attempt to balance. At best, this may be pointless, and at worst, it could be highly misleading, either masking the problem or erroneously assuming a fuel leak.

Given the above scenario, I would not be happy with an automatic crossfeed system that did not have a manual override. In which case, why have an automatic system at all.

That's a perfectly valid point. But the dangers of the crossfeed system still exists. A more ideal fuel system would, in my mind (though not in the mind of others, believe me) would include direct fuel measurement through ultrasound and capacitance, fuel useage calculation through FADEC, fuel uplift measurement through flow gauges in the valves all resulting in a more comprehensive fuel overview than at present.

Quoting BA777ER236 (Reply 65):
Yes quite Zeke:- ziplip

I don't see why suggesting that flight crews will make a mistake results in everyone acting like I've comitted a mortal sin. Of course it will happen again.

Quoting BA777ER236 (Reply 65):
You seem to think that us pilots are a constant threat to flight safety whereas with careful engineering design and thoughtful ergonomic design we can enhance flight safety from both directions. Please don't design and incorporate systems just for the sake of it though.

In this day an age everyone consults with the airlines and maintenance staff and pilots, see the result in the 777. But flight crews have to realise that there isn't some great plot by engineers with the manufacturers to get rid of them from the flight deck. The point is to ease pilot work load and improve flight safety, that's all.
Your bone's got a little machine
 
User avatar
zeke
Posts: 15265
Joined: Thu Dec 14, 2006 1:42 pm

RE: A32x Corrupt Software

Wed Jan 31, 2007 10:16 pm

Quoting 777236ER (Reply 66):
So you would prefer the situation where screens go dark, lights go out, p/bs are unlit and the restoration of power comes down to the crew pressing the AC ESS p/b - when the recent evidence has shown clear that crew sometimes won't do that?

I have flown many aircraft without electric screens and autopilots, the failure of them does not concern me at all. What does concern me is that electric systems are normally very reliable, and for them to fail could point to bigger bus related problems. I dont mind the automatic reconfiguration in the event of an engine/generator/idg failure, but an automatic reconfiguration in the event of a bus failure I do have some concern over as that maybe the only indication of a pending electrical fire.

At the same time I do not like automatic lateral crossfeed, as automation itself brings problems. One day in my view that will lead to an automatic crossfeed when a subtle leak was present going unnoticed maybe for some time. Pushing the crossfeed for me is to question what is happening, why are we doing this, and for how long ?

Quoting 777236ER (Reply 66):
That's a bit of a cop out, most crashes can be partly attributed to 'organisational failures' before flight leaving the aircraft or crew in a dangerous position.

Well in that case the company knowingly put the incorrect part on an ETOPS aircraft. Organisational failures such as supervision, training, and management were all over this incident. They paid a large fine for doing so.

Quoting 777236ER (Reply 66):
Oh well then, not worth considering is it! The fact is that bleed air leaks cause spurious engine fire warnings, your experience notwithstanding.

Well I have had more cargo fire and toilet indications, I cannot remember the last time I have read about a false bleed air fire warning. From memory Boeings reasoning for this part was not the same as yours, theirs had to do with maintenance time and cost reducing and an extension on engine life, so basically it was an accounting decision not a safety decision.

Quoting 777236ER (Reply 66):
You seem to think that matters relating to flight should be more automated, but matters relating to sytems should be less automated?

Not "less" automated per say, with sufficient manual controls.

Quoting 777236ER (Reply 66):
I don't see why suggesting that flight crews will make a mistake results in everyone acting like I've comitted a mortal sin.

A number of layers made mistakes before the pilots did, I do not see you advocating those layers to be automated. Going by your line of argument in this thread automation makes systems safer, what will make the safety system safer ?

Will automation fix cultural differences ? Why is it in some countries they can fly aircraft that are 30 years old and have next to zero incidents, when others have a terrible record ? Why do some companies have more incidents than others ?

Safety systems cannot be automated, it is the overall system where the aircraft is just one part that needs to be addressed.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
User avatar
Starlionblue
Posts: 20175
Joined: Fri Feb 27, 2004 9:54 pm

RE: A32x Corrupt Software

Wed Jan 31, 2007 11:16 pm

Quoting Zeke (Reply 67):
Will automation fix cultural differences ? Why is it in some countries they can fly aircraft that are 30 years old and have next to zero incidents, when others have a terrible record ? Why do some companies have more incidents than others ?

Safety systems cannot be automated, it is the overall system where the aircraft is just one part that needs to be addressed.

Or as I would put it, never forget those pesky fleshbags when designing a system. They have a way of screwing up even the most idiot proff equipment and procedures if you let them.

[Edited 2007-01-31 15:17:31]
"There are no stupid questions, but there are a lot of inquisitive idiots." - John Ringo
 
Pihero
Posts: 4318
Joined: Mon Jan 31, 2005 5:11 am

RE: A32x Corrupt Software

Thu Feb 01, 2007 6:51 am

Quoting Zeke (Reply 67):
A number of layers made mistakes before the pilots did, I do not see you advocating those layers to be automated. Going by your line of argument in this thread automation makes systems safer, what will make the safety system safer ?

Because it's Pilot Error

Quoting Zeke (Reply 67):
Will automation fix cultural differences ? Why is it in some countries they can fly aircraft that are 30 years old and have next to zero incidents, when others have a terrible record ? Why do some companies have more incidents than others ?

The night was dark and the captain intent on his flying. F/o and F/E saw the aircraft banking slowly, then increasingly fast to one side...Captains horizon was still level...
"Captain we're turning"
"No we're not"
"Captain, we're banking "
"Shut up ! I'm the commander..."
(Among the most famous last words )

This thread reminds me a lot of a sci-fi novel I enjoyed a lot "The Solarians" by Norman Spinrad., in which, in 1966, he foresaw the trend towards automats. The human race is at war with a very logical race bent on genocide. The solution was to erase all we knew and in fact promote the avent of a new humanity (the description of a mental calculus driven hand-flying of a starship is an absolute thrill).
Zeke, let's fly a Cat III B landing ! It's not a gimmick, I promise you !

Regards to all.
Contrail designer
 
User avatar
Starlionblue
Posts: 20175
Joined: Fri Feb 27, 2004 9:54 pm

RE: A32x Corrupt Software

Thu Feb 01, 2007 1:35 pm

Some of this is a bit above my head, so I will step back a bit and talk about the more "philosophical" parts instead of the details.

My issue with automating things easily done manually is that it assumes the pilot is the problem. It is a mistake to exclude pilots from decision loops completely. What's the point of putting the two fleshbags up there if they are not allowed to use their brains? After all, those brains are the only two objects in the aircraft systems which can actually deal with the unexpected.

Granted, sometimes the pilots will still get it wrong, but they are there where and when it happens. It is much more probable that they will make the correct decision (as is borne out by all the planes in unusual situations flying safely every day) than a systems designer who will not be in a cockpit when the fecal matter impacts the rotary air impeller.

Make the systems better, but don't exclude the fleshbags. They can be improved too. We call it training.

It is a commonly held belief among engineers that they can build a perfect machine. There is no such thing. By giving decision authority to the people on the spot, and providing the tools to enact the decisions, the system as a whole becomes safer. While some may think Airbus is not doing this, it should be remembered that the automation is simply turned off if there is a major malfunction. The aircraft then becomes just another one with traditional controls. And so the fleshbags have the tools to solve the problem, one way or another.
"There are no stupid questions, but there are a lot of inquisitive idiots." - John Ringo
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Thu Feb 01, 2007 9:33 pm

Quoting Zeke (Reply 67):
I have flown many aircraft without electric screens and autopilots, the failure of them does not concern me at all.

Why? The failure of autopilots at altitudes puts the, perhaps sleepy and reserve, crew into a situation where they're flying an aircraft at what is essentially near the extreme end of the envelope when they're not expecting it. An unexpected autopilot disengage increases the workload of both crew at a time when there are system failures. Regardless of whether 'the failure of electric screens' concerns you, I would hope you feel some sort of negative emotion towards the idea of an A320 flying around with only the lower ECAM operative.

Quoting Zeke (Reply 67):
I dont mind the automatic reconfiguration in the event of an engine/generator/idg failure, but an automatic reconfiguration in the event of a bus failure I do have some concern over as that maybe the only indication of a pending electrical fire.

Automatic reconfiguration comes with ECAM/EICAS advisories, cautions and warnings, obviously. You seem to be advocating the situations described above, where failure of the crew to adequately understand the system lead to periods of time without essential equipment - and in the case of the easyJet incident - an entire flight without RTF.

Quoting Zeke (Reply 67):
At the same time I do not like automatic lateral crossfeed, as automation itself brings problems. One day in my view that will lead to an automatic crossfeed when a subtle leak was present going unnoticed maybe for some time. Pushing the crossfeed for me is to question what is happening, why are we doing this, and for how long ?

So what's better, the consideration of engineers and pilots on the ground, months before the aircraft flies and continuous improvement through the life of the aircraft regarding potential emergencies...or getting a crew with limited knowledge of the system to trouble shoot in air? There's nothing stopping a computer with more knowledge of the system and a clearer overview of the system asking the same questions when it operates the crossfeed valve. You may be arrogant in your implication that you're 'beter' than the Air Transat crew, but remember that they were a well-trained crew with good CRM flying a modern type with all the information you would get...and they still made a fatal mistake.

Quoting Zeke (Reply 67):
Well in that case the company knowingly put the incorrect part on an ETOPS aircraft. Organisational failures such as supervision, training, and management were all over this incident. They paid a large fine for doing so.

Yes, of course that's true, but it's not relevant to the discussion. The assumption has to be that failures can happen in flight, regardless of whether they've come from maintenance problems, cultural problems, HR problems or technical problems.

Quoting Zeke (Reply 67):
A number of layers made mistakes before the pilots did, I do not see you advocating those layers to be automated. Going by your line of argument in this thread automation makes systems safer, what will make the safety system safer ?

The layers that made mistakes before the pilots did weren't on the aircraft. It's not relevant how they should be dealt with, we're talking about aircraft.

Quoting Zeke (Reply 67):
Will automation fix cultural differences ? Why is it in some countries they can fly aircraft that are 30 years old and have next to zero incidents, when others have a terrible record ?

And I could ask you why CRM-savvy, Western crews flying brand new aircraft still make silly mistakes like crossfeeding fuel into a leaking tank and not reconfiguring the electrical system correctly. Culture issues are separate from systems on an aircraft. The only interation the system has is with the two flight crew during the given even - all that happened before was irrelevent.

Your reasoning is a bit bizarre, to my mind. You advocate some level of systems automation, but still feel that pilots should have significant control to maintain situational awareness and 'deal with problems', without defining what those would be. On the other hand, you're not in favour of HUDs - perhaps the one thing in 15 years that's been a function of engineers working with crew to give them more situational awareness about the actual flight of the aircraft.
Your bone's got a little machine
 
User avatar
zeke
Posts: 15265
Joined: Thu Dec 14, 2006 1:42 pm

RE: A32x Corrupt Software

Fri Feb 02, 2007 12:57 am

Quoting 777236ER (Reply 71):
Why? The failure of autopilots at altitudes puts the, perhaps sleepy and reserve, crew into a situation where they're flying an aircraft at what is essentially near the extreme end of the envelope when they're not expecting it.

Why so melodramatic ? loss of an autopilot at altitude is not an emergency, I have hand flown an Airbus from during the climb, cruise, and descent before when the autopilots were not available, it is not as difficult as you are trying to portray. Airliners do not require the skill that say the U2 pilots had when the only had a 2kt buffer between the high and low speed buffet.

I have even landed in direct law, it is no big deal either.

Quoting 777236ER (Reply 71):
Regardless of whether 'the failure of electric screens' concerns you, I would hope you feel some sort of negative emotion towards the idea of an A320 flying around with only the lower ECAM operative.

Not really, only if that was say associated with volcanic ash, it does not take that long to run through ecam, in that time frame I am more than capable of hand flying an aircraft on standby instruments.

This is something I have done in the sim, along with engine failure, if I were to encounter either online, I would be comfortable with it as I have seen it before.

Quoting 777236ER (Reply 71):
You seem to be advocating the situations described above, where failure of the crew to adequately understand the system lead to periods of time without essential equipment - and in the case of the easyJet incident - an entire flight without RTF.

Aviate, navigate, communicate are my order of priorities, I happily go for hours at time these days without talking to people on a radio, using CPDLC instead, it does not phase me. The person on the other end of the radio is of little use to me in that sort of situation, I would contact them in due course. The transponder was working, they had 7700 set, and continued via normal flight plan, ATC did know they were coming.

In the easyjet incident it was the automation which was the problem. The aircraft dispatched with only one engine generator connected to the bus, with the APU powering the other. The GCU1 (which was known to be faulty before the flight commenced, hence the APU was online) monitoring system had incorrectly interpreted the fault in the GCU itself as a fault in the open GLC1. The GCU had consequently locked open BTC1, thus disconnecting the APU generator from the AC1 bus.

As far as I kniw, the crew did try AC ESS feed button, that did not work, a fair bit could have been recovered if they had manually hit the emergency generator, that is a backup means to get part of the AC ESS, DC ESS, DC BAT back.

I think some of the logic needs to be changed so that in the event of DC BUS1 and DC BAT being unpowered the DC1 tie contactor should open, and the DC2 tie contactor should close. That could effectively power the AC ESS via AC2, TR2, DC2, DC2 tie contactor, DC BAT, the static invertors to the AC ESS. The AC & DC ESS SHED most prob would still be lost, that is no biggie. I think the DC ESS should have the ability to be powered via DC2. That would provide another level of redundancy if the AC ESS feed button does not work.

Quoting 777236ER (Reply 71):
There's nothing stopping a computer with more knowledge of the system and a clearer overview of the system asking the same questions when it operates the crossfeed valve.

Any what exactly is going to power that computer ? The reason for the radios not working on the easyjet flight was they were of a new design off the same bus.

It was basically the fault of a computer (GCU1 monitoring system) that lead to the easyjet incident in the first place.

Quoting 777236ER (Reply 71):
You may be arrogant in your implication that you're 'beter' than the Air Transat crew, but remember that they were a well-trained crew with good CRM flying a modern type with all the information you would get...and they still made a fatal mistake.

No, not really, I just come from an airline that places safety as its highest priority, it is the first line of our corporate vision. I see incidents at other airlines which could have been prevented by ground based saftey system before the aircraft moves. I don’t believe a pilot is super human, nor do I believe aircraft design and manufacturing is infallible.

Our corporate culture is different, e.g. during a typhoon last year, the airline just cancelled flights, thereby taking the risk and pressure off pilots, that sort of decision is not cheap, but much cheaper than an accident.

Quoting 777236ER (Reply 71):
Yes, of course that's true, but it's not relevant to the discussion.

That is your view, and I can accept that, I am sure other do share the same viewpoint, which I hope you can accept.

Quoting 777236ER (Reply 71):
You advocate some level of systems automation, but still feel that pilots should have significant control to maintain situational awareness and 'deal with problems', without defining what those would be

That is a general philosophy, which could be applied to each and every system. To use your train or thought that every system should be automated, on the detection of an engine fire automatic shutoff of that engine should occur, and agents discharged. I don’t agree with that, I think pilots should control engine shutoff, and fire extinguishing.

Quoting 777236ER (Reply 71):
On the other hand, you're not in favour of HUDs - perhaps the one thing in 15 years that's been a function of engineers working with crew to give them more situational awareness about the actual flight of the aircraft.

Well they have been available as an option for many years on the Airbus FBW family, with very few people (if any) installing them. They have also been installed on 737s, with only one side having them installed.

I just have not seen a safety driven argument for their installation, please note only recently a Dash 8 ran off a runway when doing a CAT 3 landing that was equipped with a HUD.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Fri Feb 02, 2007 11:10 pm

I'm a bit too busy to be too verbose, so I'll only pick and choose what to reply to, if you don't mind.

Quoting Zeke (Reply 72):
Aviate, navigate, communicate are my order of priorities,

This is the crux of the matter. The entire point of all automation is to allow you to do those things better. None of those three priorities particularly consider the intricacies of systems - so why not let the systems take care of themselves? You have to bear in mind that automation has increased by orders of magnitude of the last few decades, and that the majority of systems changes are handled by the aircraft, not the crew.

As an engineer, I would find it better for a pilot facing an emergency to be able to aviate, navigate then communicate without worrying about the configuration of systems on board the aircraft.

Quoting Zeke (Reply 72):
it does not take that long to run through ecam,

This is another contradiction. You don't mind the system analysing the problem and giving instructions on the ECAM, which you follow, but you do mind the system doing the actions itself. What's more likely - the system to perform an ECAM action that's not appropriate, or the crew to ignore the ECAM actions?

Quoting Zeke (Reply 72):
The transponder was working, they had 7700 set, and continued via normal flight plan, ATC did know they were coming.

All very well, but this shows another crew error. Despite their training, they didn't set 7600 to show they had a radio failure.

Quoting Zeke (Reply 72):
I just have not seen a safety driven argument for their installation, please note only recently a Dash 8 ran off a runway when doing a CAT 3 landing that was equipped with a HUD.

Well using your single-incident idea you could argue that A320 FBW in unsafe because of Mulhouse-Habsheim.

HUDs go hand in hand with other emerging technology and show a change in philosophy in the flight deck. The aim now is to get crews to have greater situational awareness of the environment around the aircraft with eyes looking out of the cockpit. HUDs, enhanced vision systems, synthetic visual displays, whatever you want to call it, all mean that information that crew had to look down to get, often scattered across three or more screens, can be presented to them in a clearer manner in a way that allows them to continue looking out of the aircraft.

Imagine the evolution from a simple PFD with regular flight director to a heads up display that shows full EGPWS terrain information, with synthetic systems showing exactly where the terrain is, aircraft performance limits (speed limitations, pitch and roll limits), TCAS danger areas, visible windsheer warnings, visual, 3D synthetic navigation information all with the pilot keeping his eyes out of the window and not burried down at the displays.
Your bone's got a little machine
 
User avatar
zeke
Posts: 15265
Joined: Thu Dec 14, 2006 1:42 pm

RE: A32x Corrupt Software

Sat Feb 03, 2007 2:11 am

Quoting 777236ER (Reply 73):
As an engineer, I would find it better for a pilot facing an emergency to be able to aviate, navigate then communicate without worrying about the configuration of systems on board the aircraft.

In the case of an AC bus failure, nothing I have seen so far smells or sees smoke in the cockpit, hence my apprehension for automatic configurations. I also like the ability to manually take a bus off line if that is a source of fire.

I think I have not myself clear above to you, my concerns with automation is when it can make life more difficult for me, I can handle a degraded mode for a while, but if I get an electrical fire when I dont want one, or an engine shutting down when I dont want it to, it is where I have my concern.

Quoting 777236ER (Reply 73):
This is another contradiction. You don't mind the system analysing the problem and giving instructions on the ECAM, which you follow, but you do mind the system doing the actions itself. What's more likely - the system to perform an ECAM action that's not appropriate, or the crew to ignore the ECAM actions?

ECAM can always be overridden by the commanders judgement. Automatic reconfiguration cannot. It is what got the easyjet flight into trouble in the first place. We ignore aspects of ECAM if its appropriate, or do actions which ECAM does not even suggest, thats the commanders discretion.

Quoting 777236ER (Reply 73):
Despite their training, they didn't set 7600 to show they had a radio failure.

It was more than just a radio/communication failure, it is the commanders discretion.

Quoting 777236ER (Reply 73):

Well using your single-incident idea you could argue that A320 FBW in unsafe because of Mulhouse-Habsheim.

FBW has been around a lot longer than that, go back to the 1950s for the vulcan, and the concorde in the 1960s.

Quoting 777236ER (Reply 73):
HUDs, enhanced vision systems, synthetic visual displays

Those three technologies are not the same.

Quoting 777236ER (Reply 73):
all mean that information that crew had to look down to get, often scattered across three or more screens, can be presented to them in a clearer manner in a way that allows them to continue looking out of the aircraft.

Still need to look inside, anyone who told you different is yanking your chain.

Quoting 777236ER (Reply 73):
Imagine the evolution from a simple PFD with regular flight director to a heads up display that shows full EGPWS terrain information, with synthetic systems showing exactly where the terrain is, aircraft performance limits (speed limitations, pitch and roll limits), TCAS danger areas, visible windsheer warnings, visual, 3D synthetic navigation information all with the pilot keeping his eyes out of the window and not burried down at the displays

That is not what is going into these aircraft, that is some pipe dream years away. HUDs are very basic in what they display, no where near as sophisticated as you suggest.

http://www.blogsouthwest.com/wp-content/uploads/2006/08/WEB%20HUD%20at%20Breakout.jpg
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Sat Feb 03, 2007 3:23 am

Quoting Zeke (Reply 74):
That is not what is going into these aircraft, that is some pipe dream years away. HUDs are very basic in what they display, no where near as sophisticated as you suggest.

Airbus is working on exactly what I mentioned, I suspect that Boeing's inclusion of the standard HUD on the 787 means the same thing.

Quoting Zeke (Reply 74):
I think I have not myself clear above to you, my concerns with automation is when it can make life more difficult for me, I can handle a degraded mode for a while, but if I get an electrical fire when I dont want one, or an engine shutting down when I dont want it to, it is where I have my concern.

So why would an automatic crossfeed valve make things more difficult to you? Why would an automatic essential bus switcher make things more difficult to you? With those two specific suggestions I've made there's no degredation of any safety, and you still haven't given an example of a reduction in safety that my suggestions have made.

Quoting Zeke (Reply 74):
We ignore aspects of ECAM if its appropriate, or do actions which ECAM does not even suggest, thats the commanders discretion.

Exactly. Crew ignore ECAM all the time - in the case of the Air Transat A330 it's by the grace of God that everyone lived.

Quoting Zeke (Reply 74):
FBW has been around a lot longer than that, go back to the 1950s for the vulcan, and the concorde in the 1960s.

So have heads up displays! You're really going in circles here, it seems that you've forgotten what the original point was you were making.

Quoting Zeke (Reply 74):
Those three technologies are not the same.

They're being developed together.

Quoting Zeke (Reply 74):

Still need to look inside, anyone who told you different is yanking your chain.

During the first stages of an emergency, visual or ILS landing or take off, can you tell me why?
Your bone's got a little machine
 
User avatar
zeke
Posts: 15265
Joined: Thu Dec 14, 2006 1:42 pm

RE: A32x Corrupt Software

Sat Feb 03, 2007 5:37 am

Quoting 777236ER (Reply 75):
So why would an automatic crossfeed valve make things more difficult to you?

Because if it happens without my knowledge, I will not not why it is doing so.

The effort in turning on the crossfeed is next to nil, working out the reason for turning it on is the key. It is actually one thing I dont need to turn on for example on an engine failure after takeoff in a twin, ECAM says to do so, I take the view if I am coming back I do not need it on.

If we have a leak, I dont want the system to automatically turn it on to even up the balance. The aircraft only has a finite amount of fuel.

Quoting 777236ER (Reply 75):
Why would an automatic essential bus switcher make things more difficult to you?

To use that button, one needs to have AC1 bus fault.

If that bus fault is associated with a smell of sight of smoke, the last thing I want the aircraft to do is power the bus back up again, could cause more fire, or take out the other bus and leave me with nothing.

Do you understand the AC1/AC2 relationship when GEN1 goes offline ? The AC ESS is not used in that case.

Quoting 777236ER (Reply 75):
With those two specific suggestions I've made there's no degredation of any safety, and you still haven't given an example of a reduction in safety that my suggestions have made.

I have mentioned the same reasons above before, to me it seems like you lack operational experience to understand why that is so important to a pilot.

I also gave you the example of fire push buttons, and agents, I dont want them to be automatic either.

Quoting 777236ER (Reply 75):
Exactly. Crew ignore ECAM all the time - in the case of the Air Transat A330 it's by the grace of God that everyone lived.

It is becoming very clear to me that you do not understand the chain of events with the Air Transat A330 incident. The report is yet to be published so I will continue to refrain from explicitly telling you what and why things happened, from your comments in this thread, you are missing some fundamental understanding of what happened.

Quoting 777236ER (Reply 75):
So have heads up displays! You're really going in circles here, it seems that you've forgotten what the original point was you were making

I still maintain HUDs as in the one I have shown in my previous post to me are just gimmicks, it does not provide me with any additional information.

I would not disconnect the autopilot and hand fly an CAT 3 approach on a HUD, while it may be possible, and legal, in my view it is not very smart. I leave the autopilot in for the landing or the missed approach with Wx below CAT 1, it stays in all the time.

I am not going to go into aerial combat, so I do not need targeting of weapons, and I have full runway rollout capability even with an engine out, so it does not provide me with additional capability.

It is beneficial to aircraft that do not have full rollout capability, however I think aircraft that do not have that ability should not be doing CAT 2/3 landings in those conditions to begin with.

Quoting 777236ER (Reply 75):
During the first stages of an emergency, visual or ILS landing or take off, can you tell me why?

To monitor the aircraft during all phases of flight, even during a visual circuit we spend a lot of our effort inside the cockpit. Jets are not flown "by the seat of ones pants", it is all thrust and attitude.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
Valcory
Posts: 88
Joined: Fri Dec 22, 2006 6:56 pm

RE: A32x Corrupt Software

Sat Feb 03, 2007 6:16 am

Quoting 777236ER (Reply 55):
In the arcing example, it would be a simple solution to put a current monitor in the system that could detect arcing.

What about an On/Off switch
 
Pihero
Posts: 4318
Joined: Mon Jan 31, 2005 5:11 am

RE: A32x Corrupt Software

Sat Feb 03, 2007 8:43 pm

Quoting Valcory (Reply 77):
What about an On/Off switch

?

Are you out of your mind ?
An on/off switch for a thick fingered pilot to use without understanding it ?  Wink

Quoting Zeke (Reply 76):
I still maintain HUDs as in the one I have shown in my previous post to me are just gimmicks

It seems I've led both of you astray on a tangent here : The point I made was that, given the right tools, a pilot could do away with a lot of the automation that had been brought into the cockpits. As an example i imagined a HUD, coupled with an enhanced vision system and maybe a synthetic visual display ; that would bring the pilot's capability to a that of a visual approach in VMC. No need for a triple / quadruple cat III B autoland system, it would be cheaper to buy and maintain for the airlines...
Yes, Zeke, the HUD you showed is an incomplete tool, but for having used it on line for a few years, I can tell that it makes a pleasure of the most demanding circling-in -a mountain procedure, gives a very relaxed an accurate handling and provides - with the ILS "box" - a head-up monitoring of the final stages of an autoland progress to complete the F/O's.

Regards
Contrail designer
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Sat Feb 03, 2007 9:48 pm

Quoting Zeke (Reply 76):
Because if it happens without my knowledge, I will not not why it is doing so.

Why would you ever need to know? There are no possible scenarios where you would make a better decision over a computer with a better knowledge of the system and a better knowledge of what's going on.

Quoting Zeke (Reply 76):

If we have a leak, I dont want the system to automatically turn it on to even up the balance. The aircraft only has a finite amount of fuel.

A fuel computer system can calculate the amount in a tank, the fuel use from FADEC, the expected amount in the tank, and work out - more accurately than you could - if there's a leak. We've already seen what happens when flight crews are faced with a leak.

Quoting Zeke (Reply 76):

I still maintain HUDs as in the one I have shown in my previous post to me are just gimmicks, it does not provide me with any additional information.

But they provide you with greater situational awareness, which you've been saying you want from the start.

Quoting Zeke (Reply 76):
To monitor the aircraft during all phases of flight, even during a visual circuit we spend a lot of our effort inside the cockpit.

Exactly. Automation of systems frees up the crew for more important things, like keeping their eyes out of the cockpit during a visual circuit.

Quoting Valcory (Reply 77):

What about an On/Off switch

It's another system for crew to foul up.

Quoting Pihero (Reply 78):

It seems I've led both of you astray on a tangent here : The point I made was that, given the right tools, a pilot could do away with a lot of the automation that had been brought into the cockpits. As an example i imagined a HUD, coupled with an enhanced vision system and maybe a synthetic visual display ; that would bring the pilot's capability to a that of a visual approach in VMC. No need for a triple / quadruple cat III B autoland system, it would be cheaper to buy and maintain for the airlines...

Ah, but you're talking about different automation now. You're talking about autoflight systems, which for all their benefits can reduce the flying skills of the crew. HUDs go hand in hand with automated systems so that crew can keep their eyes out of the aircraft.
Your bone's got a little machine
 
David L
Posts: 8551
Joined: Tue May 18, 1999 2:26 am

RE: A32x Corrupt Software

Sun Feb 04, 2007 1:26 am

Quoting Pihero (Reply 78):
An on/off switch for a thick fingered pilot to use without understanding it ?

 biggrin  ... with a big label that says "DO NOT EAT!".

Quoting Zeke (Reply 76):
Quoting 777236ER (Reply 75):
So why would an automatic crossfeed valve make things more difficult to you?

Because if it happens without my knowledge, I will not not why it is doing so.

It did occur to me that operating a switch might make the crew more aware of a problem than just letting the problem "take care of itself".
 
777236ER
Posts: 12213
Joined: Sat Aug 18, 2001 7:10 am

RE: A32x Corrupt Software

Sun Feb 04, 2007 8:13 am

Quoting Zeke (Reply 80):

You mean like when we dispatch with it U/S ? Or cabin crew or passengers tell us of a visual report of a leak ?

I don't understand why those points would preclude an automatic system. Anyone can easily make an automated system unservicable, and you don't get the risk of crew resetting the c/b in flight or trying to crossfeed when there's a fault with the system. If passengers or cabin crew can see a leak then a fully integrated fuel systems monitor would be seeing one too.

Quoting Zeke (Reply 80):
I am yet to see a fuel computer system that 100% reliable.

And I'm yet to see a pilot that's 100% reliable. That's not an insult against pilots (I have yet to see a 100% reliable engineer!), nor is it an attack against you, but you have to accept that a last amount of aircraft crashes are caused by crew interaction with systems.

Take the crash of AA587. You, and the vast majority of pilots, might realise that stomping on the pedals repeatedly at 300kts or whatever airspeed they were going, is a bad idea and is going to overstress the fin. But it takes just one pilot to do that and down the aircraft.

Now from our point of view as engineers, we have two choices. We can either go down the route of putting the relevent information in flight manuals and training programmes, so that crew are aware of the system, its limitation and why, for example, you shouldn't perform repeated full-deflection rudder movements.

On the other hand, we can design a system that looks all the pertinent flight information - airspeed, altitude etc - to work out the dynamic loads that the fin will experience, then build into the system a limit to how many reversals a rudder will perform in a given period of time. It's not a difficult program, and it doesn't even need FBW to acomplish.

Which option is better? Yes, the first option might cost less, 'keep pilots in the loop' and generally look the most sensible. But how many lives would be saved if the system existed?

Don't assume that I think the philosophy of automation only applies to what the flight crew do. I think it should be applied throughout the board - including things like maintenence. Interestingly, you could say that the engineering profession has already been through the automation, with so much stuff being done on Matlab, in FE packages, in FV CFD packages, in CATIA etc.

Edit: And also, the engineering profession is facing the same problem as flight crews with automation. Without sounding too old, so many graduates come out of universities nowadays who know that turbulence 'means you turn on the K-Epsilon model', but not a lot else.

Regarding HUDs and EVSs, check out the Thales website. They're doing a lot of work with Bombardier on the Global Express EVS (which might already be certified? I don't know), and they're also doing the A380 LCD HUD and the A400M IR HUD. A terrible supplier and company to deal with, if anyone's interested, but they tend to get the job done and are at or near the state of the art, ahead of people like Rockwell Collins and Honeywell.

[Edited 2007-02-04 00:17:06]
Your bone's got a little machine
 
Pihero
Posts: 4318
Joined: Mon Jan 31, 2005 5:11 am

RE: A32x Corrupt Software

Tue Feb 06, 2007 6:10 am

I finally managed to find this article:
here

I love the title : "The cognitive consequences of patterns of information flow " by Ed Hutchins, U of California, San Diego.
An excerpt of the article sums up quite nicely what pilots have been thinking all along :
-"In general, we should not use automation to make decisions for pilots. Rather we
should use automation to present pilots with representations of decision spaces in
which conceptually meaningful distinctions in the decision space are represented by
perceptually meaningful distinctions in the appearance or behavior of the display.
In such systems, pilots can maintain their engagement with the task of flying and
operating the airplane in an embodied rather than disembodied manner. We should
be building environments for action in which pilots get to be smart by doing what is
easy for them to do. In this way we could bring pilots back into the loop and make
them operators in a meaningful world of action rather than supervisory controllers
of a fully automated process."


Anything else is not acceptable, in terms of safety.

Regards
Contrail designer
 
User avatar
Starlionblue
Posts: 20175
Joined: Fri Feb 27, 2004 9:54 pm

RE: A32x Corrupt Software

Tue Feb 06, 2007 6:51 am

Quoting 777236ER (Reply 81):

Take the crash of AA587. You, and the vast majority of pilots, might realise that stomping on the pedals repeatedly at 300kts or whatever airspeed they were going, is a bad idea and is going to overstress the fin. But it takes just one pilot to do that and down the aircraft.

This a bad example. On any Airbus apart from 300/310, the electronics would have prevented the maneuver.

The point of the limitation is to prevent the pilots from inadvertently overstressing the aircraft, while simultaneously giving the max maneuvrability. If the stomp on the pedals to the max the aircraft will deliver max maneuver capability while staying inside break-up limits.

That's "good" automation since we are talking about time critical issues when the pilots should not be hampered with worrying if the fin is going to come off.


However, automating non time critical things like crossfeed valves just opens up the door for a whole new range of errors.
"There are no stupid questions, but there are a lot of inquisitive idiots." - John Ringo

Who is online

Users browsing this forum: drdisque, marcory, RL777, wetwing and 16 guests

Popular Searches On Airliners.net

Top Photos of Last:   24 Hours  •  48 Hours  •  7 Days  •  30 Days  •  180 Days  •  365 Days  •  All Time

Military Aircraft Every type from fighters to helicopters from air forces around the globe

Classic Airliners Props and jets from the good old days

Flight Decks Views from inside the cockpit

Aircraft Cabins Passenger cabin shots showing seat arrangements as well as cargo aircraft interior

Cargo Aircraft Pictures of great freighter aircraft

Government Aircraft Aircraft flying government officials

Helicopters Our large helicopter section. Both military and civil versions

Blimps / Airships Everything from the Goodyear blimp to the Zeppelin

Night Photos Beautiful shots taken while the sun is below the horizon

Accidents Accident, incident and crash related photos

Air to Air Photos taken by airborne photographers of airborne aircraft

Special Paint Schemes Aircraft painted in beautiful and original liveries

Airport Overviews Airport overviews from the air or ground

Tails and Winglets Tail and Winglet closeups with beautiful airline logos