Why WON'T Planes fall from sky?

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

Will someone please explain why non-compliant computerized control systems and embedded chips in larger aircraft DO NOT pose a threat to the airworthiness of the aircraft? It is my understanding that none of these planes have any manual(mechanical) back up controls, but instead rely on secondary ELECTRONIC controls, which may be subject to the same Y2K vulnerability as the primary system. So, please, tell me, what is so ridiculous about the prospect of fearing uncontrollable planes falling from the air?

-- Ann Fisher (zyax55b@prodigy.com), November 30, 1998

Answers

Ann,

Very good question. I've asked myself the same thing.

The best answer IMHO is that if there is a safety question, government will not allow these planes to fly, and insurance companies will not insure flights for these planes.

Could planes fall if they were in the air at midnight on January 1, 2000? That's the way it looks to me.

Comments?

-- Kevin (mixesmusic@worldnet.att.net), November 30, 1998.


Anne,

Last I checked, NO commercial aircraft of any size or description use "fly-by-wire" or electronic-driven controls as their primary control system. Only a few military types are primarilly fly-by-wire. Even the largest 757 can be flown safely in the absence of electric power and computer controls; even landed safely using manual hydraulic backup systems.

The risk to civil aviation lies primarilly in ground support and guidance systems. The safety of civil aviation will be secured by the insurance companies, who will disallow coverage of commercial operations should they assess undue risk.

So, contrary to the uninformed straw-canard, aircraft will not fall from the sky. They will not be allowed to leave the ground.

Hallyx

"If God had meant man to stay on the ground, He would have given him roots." (Or a mandatory insurance policy--H)

-- Hallyx (Hallyx@aol.com), November 30, 1998.


To all concerned,

Having had this discussion with many of my collegues I would add the following. Although it is true that the aircraft can be navigated and piloted in light of an electronic failure, the problem is much more complex and has deeper layers to be examined. For example, the maintenence systems of companies may have a scheduling system that handles the appointed replacements of hydraulic fluids or schedules tests on comm and navigations at certain times. These aircraft are exceedingly complex and require computers to be maintained. It is not inconcievable that because of forward or backward looking dates embedded in the scheduling system records that may exist even now that oversights in the maintenence can begin to occur. Even if the system is remediated the forward looking data which may have been created by maintenence routines performed in say, 1997 would be carried into the future, will cause failures for years to come. In short the millenium bug is already here and we will be feeling its effects in spite of remediation efforts (it has already been corrupting data for some time now). In the case of the aircraft sometime several months from now when the viscosity of a critical fluid that went unreplaced breaks down (or seals not inspected allows a leak for that matter) and the hydraulics cease to function, the plane will fall from the sky. Only it won't seem like the millenium bug because inspectors wont find an electronic failure or computer problem. They will find that a leaky seal contributed to the failure. Insideous isn't it..

-- (Thinker@TheBrain.com), November 30, 1998.


Very good question. Although I'm sure that the posted information is correct, I do think that there is a significant risk of a plane "falling out of the sky." Perhaps it would be more accurate to say, "flying into the ground." Although commercial airliners can be flown manually, they were never intended for that purpose. I would imagine, with significant guidance from the ground, providing the needed data, it could be done. I think that we all expect that ground control will be in a state of major disarray and will be unable to provide the needed assistance to the planes in the air. A few airports recently tested some new equipment and problems left 20 mile gaps in their radar coverage and also created "ghost" planes on the radar displays. (I'll try to find the URL for the data, any help would be appreciated.) The other important factor to consider is how seriously our government will take the potential security risk to the airlines. The FAA feels pretty good about itself despite their spotty performance in preparation. In light of the recent news of the DOD falsifying readiness reports, I have little faith that the agencies will do the right thing. Our government is so concerned about spin and political slap fights that they may take the chance. We want our populace to feel good don't we? We've got to "build a bridge into the next century."

K. Trout

-- Kilgore Trout (ktrout@fakeurl.com), November 30, 1998.


Last I checked, NO commercial aircraft of any size or description use "fly-by-wire" or electronic-driven controls as their primary control system.

Except for Airbus.

-- Richard Dale (rdale@figroup.co.uk), November 30, 1998.



NOT trying to undermine the general understanding that they ARE NOT going to "fall out of the sky" I would ask if the 777 is not a fly by wire ship?

I have several relatives who fly commercially and militarily, and none of them are concerned over flying into the hard stuff. They ARE concerned about the econ effects and teh cultural effects, however.

Chuck

-- Chuck a night driver (rienzoo@en.com), November 30, 1998.


767, 777 are very strongly fly-by-wire - not exclusively, but they are completely digital in the cockpit display areas.

On the other hand - Boeing says the aircraft are compliant. but I strongly question the ground power, controls, distribution, baggage handling, ticketing, scheduling, flight crew scheduling, landing/ramps assignment, refuelig, meals, maintenance (as mentioned), etc.

-- Robert A. Cook, P.E. (Kennesaw, GA) (cook.r@csaatl.com), November 30, 1998.


Two reasons why not.

the first - technical - is that unless thre is a good reason for the safety-critical avionics to know what date it is, they won't. The right way to do (say) an engine-control system is to measure time in seconds or milliseconds since the engines were started, and reset time to zero every engine-off. Overflow is either something that CAN'T happen (because the fule runs out first) or something that ALWAYS happens (say every five minutes, and therefore thorougly debugged before the bird ever flies.

The second - non-techical - is that you'll have to persuade a pilot that it's safe to take off (and an insurance company and an airline boss as well). Given various possibilities of air-traffic control failures, power blackouts, riots ... methinks airlines will ground all their planes before the fateful date and their pilots will if the airlines won't.

They'll be back in the air when it's clear that some reasonable degree of safety and normality exists again, and no sooner. Civilian pilots can't eject.

-- Nigel Arnot (nra@maxwell.ph.kcl.ac.uk), November 30, 1998.


Good thoughts all. I think the more recent Airbus models are fly-by-wire (A319, A320 etc. I subscribe to Aviation Week & Space Tecgnology which is THE trade magazine for aerospace concerns. ives great details on new developements, crashes, deals etc. Virtually NO evidence they are even aware of Y2K. Very strong silence.

I believe I read on c.s.y2k that the FAA is committed to replacing all its regional IBM 3083 mainframes with new gear. Imagine that!!!

-- RD. ->H (drherr@erols.com), November 30, 1998.


did some independent checking on this and came up with a couple of interesting bits in informal conversations - take them FWIW:

1. the planes wont fall out of the sky because the ones that have problems flying wont get into the sky - things to watch for that may not be caught in remediation but *will* show up shortly after 0001 01/01/00:

a. multiple integrated systems required to maintain balanced thrust and vectoring on multiengine turbojet aircraft - watch for at least a few instances of aircraft that attemt to taxi and suddenly swerve or veer off the taxiway...worst case - runaway turbine that just keeps going faster and faster and etc until the fuel supply is cut off or it tears itself off it's mount.

b. unexplained failures in initial startup checks of fly-by-wire aircraft- sporadic or nonresponsive control surfaces, and etc.(just pray that you have an aircrew that takes their startup checks seriously that day!). Things to watch for: aircraft experience sudden retraction of landing gear; flaps or stabilators stuck in odd positions, thrust reversers stuck in full deployed position, that sort of thing.

and then of course there's the other issue

the planes themselves wont fall out of the sky, but there's no certain way of knowing whether or not air traffic controlers responding to faulty data from badly remediated (or unremediated) systems may attempt to vector your flight through that substance pilots sometimes refer to as "cumulo-granite" (a.k.a. a cloud enshrouded mountain).

actually, I think the FAA administrator should prove she means what she says by flying to Kuala Lumpur on December 31 and then flying back on 01/01/00...

Arlin Adams

-- Arlin H. Adams (ahadams@ix.netcom.com), November 30, 1998.



Best indication we have (other threads, some national stories) is that htey replaced the programs (beahps hardware too) at several airports - San Diego, Denver, Chicago - and had major "loss of control"/"loss of data" accidents at each. New program removed. Next step - maybe - test it first?

My opinion - fly to Lampur, return a year later. If at all.

-- Robert A. Cook, P.E. (Kennesaw, GA) (cook.r@csaatl.com), November 30, 1998.


Boeing says there are sufficient "redundancy systems" so that an overlooked Y2K problem in its aircraft should pose no serious risk.

Some months ago, the Civil Aviation and Safety Administration in Australia said that Y2K risks to aircraft safety were indeed real, citing threats specifically to communications, navigation, and engine controls. Perhaps the Aussies don't know what they are talking about.

-- Don Florence (dflorence@zianet.com), December 02, 1998.


The following is a free Y2K alert + analysis from Y2KNEWSWIRE. COM.

L PLANE ACTUALLY FALLS FROM SKY
Y2K GLITCH CAUSES FIRST AIR TRAFFIC FATALITY

If the following reader e-mail turns out to be true, Y2K has claimed its first life. Not only that, the most oft used phrase by Y2K skeptics to ridicule those who understand the problem -- "Planes won't fall from the sky!" -- has turned out to be eerily on the mark. A plane *did* fall from the sky.

And who's to blame? It looks like the FAA -- precisely the agency Y2KNEWSWIRE has been accusing of irresponsible behavior for months. The FAA's repeated denial of obvious Y2K problems in the radar control software is simply unacceptable. And now, the FAA is the likely culprit for the death of this pilot (see below). How many more passengers and pilots will die while the FAA continues to beta test non-compliant radar control software on the unsuspecting public?

Note: The following, being anecdotal, hasn't been proven. We do not have news sources on this. But we have now received e-mails from three different sources about Y2K-related problems at the San Francisco airport. We believe their stories.

Here are the details: (actual e-mail from a user)

"[Last week my cousin died] in California. He was flying in a small plane, and attempting to land at SFO [San Francisco airport]. He made two attempts to land, and they told him "We've lost you" both times. He was then diverted to land at Lodi airport after the second attempt. In transit to Lodi, he ran out of gas, and the plane crashed. He died November 18th (the accident was November 3rd) at the University of California, Davis Medical Center, in Sacramento, Calif. He died from infections from severe burns sustained in the crash. He had second and third degree burns over 40% of his body. He also was being treated for a collapsed right lung, a broken jaw, a broken pelvis, and a broken back. His name was [Name withheld for privacy], a 58 year old retired professional rodeo cowboy."

Another actual e-mail about problems in the San Francisco airport:

"I work for a publishing organization here in Idaho. We were in Southern California for a Y2K Conference we held at the UCI Bren Events Center in Irvine. The conference went well and we left from LAX 10:30 a.m. to return to Spokane, WA by 3:30 p.m. on Sunday, November 29th. What happened we are still baffled with -- but we do know that Y2K upgrades were the culprits. It took my employer and I twenty-one hours to get home. We arrived in Spokane at 5:00 a.m. on Monday, November 30th. There were announcements made in San Francisco that new Y2K compliant computer upgrades were to blame for many of the problems. We were stuck there for 16 hours and many of the EXCUSES were "computer glitch" and "Y2K problems." It seemed that there was trouble knowing what planes were where, what gates were available, what the schedule was, where the planes were originating from and when they had taken off. I scanned the airport--being stranded there for so long, and listened to conversations. One woman was rattled as her plane had circled the airport so long it had to make an emergency landing out on a deserted part of the airport. There was much confusion and chaos. Even the ARRIVAL and DEPARTURE monitors were scrambled and they asked us not to rely on the information being displayed."

DO YOU BELIEVE THE STORIES?
If you do, you now realize the FAA is to blame for at least one Y2K-related death. And that goes to prove the skeptics were wrong: Y2K *will* kill some people. It already has. Furthermore, at least one plane has already fallen from the sky. And this is still 1998. Interestingly, the plane didn't fall from the sky due to Y2K failures in the airplane itself -- it fell from the sky as a result of Y2K-related failures on the ground. This is one possibility the Y2K skeptics almost universally overlooked.

If you don't believe these stories, that's fine. As we said, we can't prove them. These are anecdotes. We usually don't report anecdotes, but these were particularly important to get out to the public in order to warn people about the dangers of air travel. Given the multitude of dangerous radar failures we've reported -- and verified -- over the last few weeks, these stories seem highly credible. Remember, just last week one air traffic controller described the radar outage as, "Sheer terror," and mainstream newspapers in Chicago, Dallas, and Washington D.C. have reported on dangerous radar outages due to the FAA's software.

AVOIDING AIR TRAVEL COMPLETELY
Because of these problems, the people at Y2KNEWSWIRE are boycotting air travel effective immediately. The potential for air traffic-related problems is simply too great. We won't fly again until well after January 1, 2000. In the mean time, we predict that *hundreds* of passengers will die over the next 18 months as a result of Y2K-related glitches. Don't expect them all to be admitted Y2K problems, of course. They'll find another "explanation" for them, anything to distract attention from the FAA's looming Y2K problems.

If you want to be added to the subscription list (free!), visit http://www.y2knewswire.com and enter your e-mail address in the sign up box located at the upper-left corner of the page.
xxxxxxx xxxxxxx

-- Leska (allaha@earthlink.net), December 02, 1998.


This isn't an answer to anyone's question. This is just the latest media article I've seen on planes, flying and Y2K. Here's the link:

http://cnn.com/TECH/computing/9812/01/faa.y2k/

"Flying the unfriendly skies with the Y2K bug"

-- Kevin (mixesmusic@worldnet.att.net), December 02, 1998.


Fly-by-wire aircraft won't lose control and fall from the sky, because the software and hardware aren't based on x86 processors and Y2K-failure prone software. Instead the magic date here is 2038, when Unix-derived OS's and the processors they run on trip their clock registers. Now as for a blind air traffic control system running planes into each other... Check Six! WW PS; I work in product engineering and test at a VERY large aerospace firm (initials LM), on fly-by-wire systems and aircraft computers.

-- wildweasel (vtmldm@epix.net), December 03, 1998.


http://www.tampabayonline.net/news/news101m.htm

FAA Looking Into Planes' Near-Collision

Dec 5, 1998 -

BOSTON (AP) - Federal investigators are looking into an apparent computer mishap that nearly caused a Northwest Airlines jet to collide with an Air Ontario flight, The Boston Globe reported today.

Tuesday's close call occurred at about 22,500 feet, 30 miles southwest of Albany, N.Y., the Federal Aviation Administration said.

Only the intervention of an air traffic controller in Nashua, N.H., prevented the Northwest flight, bound from Detroit to Connecticut's Bradley International Airport, from hitting the Air Ontario commuter plane, which was flying from Providence, R.I., to Toronto, FAA officials told the newspaper.

An onboard computer system known as TCAS, for Traffic Collision Avoidance System, appears to have put the planes on a collision course.

FAA officials said the two planes flew less than a mile apart horizontally and 300 feet vertically. Adequate separation between planes in that airspace is 5 miles horizontally.

FAA spokesman Jim Peters said an investigation will include a review of the computer system, which is designed to prevent midair collisions.

The incident began when the TCAS on board Air Ontario Flight 1335, a de Havilland Dash 8-100 with 21 passengers and crew, warned of a US Airways jet taking off from Albany and heading for Pittsburgh.

Although the US Airways Boeing 737 posed no apparent threat to the Air Ontario flight, the TCAS saw only a climbing plane and ordered the Air Ontario crew to climb.

At the same time, Northwest Airlines Flight 1186, an Airbus A-320 carrying about 60 people, was ordered by its TCAS to dive.

The planes then flew so close that the targets on radar scopes in Nashua merged into one, a controller told officials.

Northwest spokesman Jon Austin said the airline was not aware of the incident, and the crew had filed no report.

Copyright 1998 Associated Press. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.

-- Leska (allaha@earthlink.net), December 05, 1998.


Moderation questions? read the FAQ