Ask The Pilot

[jb747]

Re: General Qantas Delays/Cancellations/etc. Discussion

It's a bit of a simplistic diagram. As an example, the laptops in the coughpit have to be physically switched from one system to another...not a software switch but hardware.

 

[opusman]

Events like AF447, AirAsia, Aseana...will all become more common, as the more senior pilots retire. There's a whole generation out there who are systems operators, but not pilots. It's why I'm so in favour of pilots from GA, military, low level airlines...they all know how to fly without the fancy toys.

You don't think the tech will continue to get better to counter the loss of experience?

 

[jb747]

You don't think the tech will continue to get better to counter the loss of experience?

The tech is a large part of the problem. When it runs out of ideas, it invariably has the aircraft in a situation that most pilots have never seen. The vast majority have never even been upside down.

Let's take the AF case. The pilot pitched the aircraft up (who knows why), and the speed decayed. The 'tech' helped out by trimming nose up as well, so that when it really ran out of ideas and dropped to direct law, full nose up trim had been applied. That basically means that the aircraft was now configured to hold itself in the stall...and even full forward stick may not have had enough authority to override the trim (the stick just uses the elevators, the trim uses the entire tailplane). So now, not only does it need full nose down stick to try to break the stall, but it also needs the trim to be manually driven out of that nose up setting....and that requires the use of a switch that is NEVER used in normal operations, and which isn't all that easy to quickly find. In years past, where you had to trim yourself, the trim switch lived under your thumb, but, as the tech is supposed to take away the need (as it does 99.9% of the time) this vital switch is now hidden away on the centre console.

If it gets to the point of not needing experience, you also don't need pilots. It will get there...but not for a while yet. And when it does, it will be driven by the desire to get rid of pilot salaries, not to make things safer.

 

[straitman]

You don't think the tech will continue to get better to counter the loss of experience?

opusman,

Have a look at this American Airlines video and it will further explain jb747's answers regarding automation and pilot versus systems operators. It is about 25 min long and quite comprehensive.

Whilst is is 18 years old it was very relevant then and possibly more so now.

Children of Magenta

 

[jb747]

Brilliant video which is even more true now....with the addition that for some people ending up at the lowest level of automation is an emergency in itself.

 

[opusman]

Have a look at this American Airlines video and it will further explain jb747's answers regarding automation and pilot versus systems operators. It is about 25 min long and quite comprehensive.

I don't doubt anything jb747 says, he knows far more about this than I do. But technology continues to improve - e.g. self driving cars look like they will actually be a reality soon, rather than the science fiction most people would have thought not even 3 years ago. There's no reason to think aircraft technology won't continue to improve as well. As long as the improvements in technology continue at the same rate as the loss of experience it should be a zero sum game

I'm sure accountants hate hull losses as much as they hate pilot's salaries...

 

[Boris spatsky]

Ahh Children of the Magenta. I first saw that on a Human Factors facilitators course years back. Still so relevant.

 

[Boris spatsky]

I don't doubt anything jb747 says, he knows far more about this than I do. But technology continues to improve - e.g. self driving cars look like they will actually be a reality soon, rather than the science fiction most people would have thought not even 3 years ago. There's no reason to think aircraft technology won't continue to improve as well....

Except that when the automation fails, or does something completely unpredictable, you can't just pull over and shut the engine(s) off...

 

[jb747]

I don't doubt anything jb747 says, he knows far more about this than I do. But technology continues to improve - e.g. self driving cars look like they will actually be a reality soon, rather than the science fiction most people would have thought not even 3 years ago. There's no reason to think aircraft technology won't continue to improve as well. As long as the improvements in technology continue at the same rate as the loss of experience it should be a zero sum game

I'm sure accountants hate hull losses as much as they hate pilot's salaries...

Firstly the accountants (aka management) all work on the theory that they won't be there when it goes wrong. If, in fact, they even think that much about it. I very much doubt that they equate their decisions with reducing safety.

This video is of a talk that is related to the 757 and 767. They were very much first generation glass jets. The technology of the A380 is vastly beyond them. But, the issue has become worse, not better. In part, because said accountants have been told that this wonderful tech allows the use of less qualified pilots. And, the result is that they pretty much have to live at the highest level of automation, because they don't have the knowledge to operate at the lowest. Aseana...a perfectly healthy 777 flown into the ground on a lovely visual day.

 

[opusman]

Except that when the automation fails, or does something completely unpredictable, you can't just pull over and shut the engine(s) off...

The implication is that as it improves it becomes less likely to fail

 

[jb747]

The implication is that as it improves it becomes less likely to fail

That has not been the experience with these systems at all. As their complexity has increased, unexpected results come to the fore.

Eventually perhaps. Not any time soon, Google cars notwithstanding.

 

[RAM]

That has not been the experience with these systems at all. As their complexity has increased, unexpected results come to the fore.

Eventually perhaps. Not any time soon, Google cars notwithstanding.

In Finance a common saying is "Right in theory, broke in practice." Spalvins knew that the 87 crash was coming, he was adamant and started shorting the market in late 1986. After he had lost over $130m shorting the market - he gave up around April 87 (I think it was). He ended up losing nearly everything he had built up once October 87 struck.

Technology is a wonderful tool right up until the moment that something goes wrong. So redundant systems are built in to create a safety buffer. However all that tech is maintained by an error ridden back-up system involving people who come under the control of 'senior mgmt'.

The numbers people work out that they can save $xx_ by cutting the maintenance down time by y% and outsourcing some maintenance etc. They've run the figures and the Monte Carlo simulation suggests that it only increases the probability of hull loss by 0.01% while increasing EPS by 4% which should ensure the senior mgmt team will get 100% vesting of their performance bonus.

Guess what decision they make?

Or another way of thinking about it is in reference to a patient who is being prescribed 7 or 8 different drugs. There has never been a large scale optimised study into the impact of combining drug A with drug BG1 and drug HD2 and drug YG with drug WE etc. Yet each drug in isolation has been approved. True, there have been some studies about combining one or two or three drugs to treat ONE PARTICULAR ailment. More often than not - that study was funded by the pharma that produces each of those three drugs - in the hope of being able to expand their available prescription market.

Who'd guessed that combining those 7 or 8 or 9 drugs would lead to .....


So technology is great but it is a tool not the driver. The air crew need to be able to operate in the absence of it because $%^T happens. Where they are up to the challenge then we get to have some 'champions' to acknowledge. When they're not then the industry has fingers to point.

 

[opusman]

So technology is great but it is a tool not the driver. The air crew need to be able to operate in the absence of it because $%^T happens. Where they are up to the challenge then we get to have some 'champions' to acknowledge. When they're not then the industry has fingers to point.

The pilot isn't up there flapping his arms you know. The whole concept of air travel is predicated upon trusting technology. I'm just pointing out it's not necessarily a given that you can only trust it up to a point - or if you can, that point is constantly moving forward.

 

[juddles]

At the moment my main objection to greater (even total) automation of planes is due to the tremendous influence that terrorism is enjoying, and the unbelievable billions of dollars being spent to fund it. (Yes, fund it, not fight it)

A "crazy" at the moment can work out a way to take control of and crash a plane or two. Once there are no physical pilots in them, that "crazy" can work towards crashing a thousand planes simultaneously.

 

[russ]

Re: General Qantas Delays/Cancellations/etc. Discussion

Thanks, but doesn't that mean that the laptop can pick up a virus from one system and then when you flick the switch, put the virus on the other system?

 

[moa999]

Interesting article here about substantial hail damage on a Delta 747:
Hail pummels Delta's N664US Boeing 747; NWA's "Spirit of Beijing" may face scrapyard (Photos) - Minneapolis / St. Paul Business Journal

Reportedly when flying over China the pilot asked for permission to vary course and was denied.... and this resulted.

I remember jb747 previously stating that with ATC you can require certain runways etc, even if initially given a different runway

Is this also the case on a pre-planned course/ air-lane??
Or are there certain regions (eg. China with all its military airspace, and potentially over Russia/ Middle East) where you just wouldn't and instead fly through the weather.

 

[ChrisGibbs]

JB747 - a couple of weeks back I was on the QF1 service from DXB-LHR and about 45 minutes out of LHR. There was a whole bunch of traffic flying above and below as well as what appeared to be at the same flight level converging on various ports around London (Gatwick, Stansted as well as Heathrow). During daylight flights across the Atlantic I've also seen what appears to be a whole bunch of aircraft travelling in similar directions at different and similar flight levels. It got me thinking that most commercial airliners are now equipped with GPS increasing the accuracy of navigation. With the increased volume of traffic and increased accuracy of navigation does it also follow that the probability of a collision also increases - especially where two aircraft on similar headings (operating at the same odd or even flight level) are converging around waypoints that are not under radar coverage (for example Trans Atlantic, Trans Pacific flights)? Is there some means by which the loaded flight plan has an offset or bias applied to the standard track (+/- 1nm for example) to reduce the probability of collision? If so is this something that is dictated by the company or something that is up to the PIC?


Thanks in advance...

 

[Orca]

Interesting article here about substantial hail damage on a Delta 747:
Hail pummels Delta's N664US Boeing 747; NWA's "Spirit of Beijing" may face scrapyard (Photos) - Minneapolis / St. Paul Business Journal

Reportedly when flying over China the pilot asked for permission to vary course and was denied.... and this resulted.

I remember jb747 previously stating that with ATC you can require certain runways etc, even if initially given a different runway

Is this also the case on a pre-planned course/ air-lane??
Or are there certain regions (eg. China with all its military airspace, and potentially over Russia/ Middle East) where you just wouldn't and instead fly through the weather.

Hi moa99,

For context, I'm an Australian air traffic controller - so reasonably qualified to answer your query. A pilot (specifically the Captain, when referring to RPT operations) has the ultimate responsibility for the safety of an aircraft and, by extension, its occupants. If a pilot believes the safety of flight is in question, he/she is well within their rights to take action as necessary to maintain safety. In some circumstances this will mean disregarding ATC instructions.
With regards to the scenario described, ATC may not have been able to approve the weather deviation request for any number of reasons including proximity to other aircraft or military restricted airspace. At that point, the pilot then have to assess whether deviating around the weather, without ATC clearance, is more or less of a risk to the safety of flight than going through said weather.
For interests sake, this occurs regularly during the storm season here in Australia whereby aircraft will request deviations that I am unable to issue immediately - most of the time the crew elect to remain on track until I'm able to issue said clearance but I have also had instances whereby the crew elect to deviate anyway and proceed into active MIL areas rather than track into the weather. A good controller will always provide advice to the pilot as to why the clearance is not available so they are able to make a best judgement decision. If they elect to proceed without clearance, ATC have set procedures to ensure that safety is maintained to the best of their ability.
I hope that answers your query.
Cheers

 

[Orca]

JB747 - a couple of weeks back I was on the QF1 service from DXB-LHR and about 45 minutes out of LHR. There was a whole bunch of traffic flying above and below as well as what appeared to be at the same flight level converging on various ports around London (Gatwick, Stansted as well as Heathrow). During daylight flights across the Atlantic I've also seen what appears to be a whole bunch of aircraft travelling in similar directions at different and similar flight levels. It got me thinking that most commercial airliners are now equipped with GPS increasing the accuracy of navigation. With the increased volume of traffic and increased accuracy of navigation does it also follow that the probability of a collision also increases - especially where two aircraft on similar headings (operating at the same odd or even flight level) are converging around waypoints that are not under radar coverage (for example Trans Atlantic, Trans Pacific flights)? Is there some means by which the loaded flight plan has an offset or bias applied to the standard track (+/- 1nm for example) to reduce the probability of collision? If so is this something that is dictated by the company or something that is up to the PIC?


Thanks in advance...

Hi Chris,

I hope you don't mind me answering your query - it's as much an ATC question as it is one for pilots. The nuts and bolts of it are that your summation that as air traffic volume increases so does the risk of collision are somewhat correct. However, it is for this reason that we (ATC) exist. The absolute core responsibility of an air traffic controller is to "prevent collisions between aircraft". As such, separation standards are designed specifically for the type of traffic that they will be applied to. They take into account equipment available, be it ground equipment such as radar & ADS-B or navigational capabilities onboard the aircraft. The separation standards we use vary - for instance, between SY and BN we apply 5 nautical miles radar sep between aircraft at co-altitudes. However flying to NZ, we apply a minimum of 10 minutes between aircraft at the same altitude (generally, there are exceptions). My point being that aircraft are positively separated at all times during all phases of flight within controlled airspace (again, there are caveats but you get the gist).
In Oceanic airspace throughout most of the world pilots are allowed to off-set laterally from their designated route up to 1 or 2 nautical miles to the right of track if they so desire. As I understand it, most companies allow the pilot to make that call but there are a few that mandate the use of such a procedure.
I hope that helps you

 

[jb747]

Re: General Qantas Delays/Cancellations/etc. Discussion

Thanks, but doesn't that mean that the laptop can pick up a virus from one system and then when you flick the switch, put the virus on the other system?

I didn't say that very well. The laptop is always isolated from the aircraft system. All we are switching is our access...i.e. the keyboard and display.