MK Airlines B747 crash at Halifax
Join Date: Oct 1999
Location: New Delhi, India
Posts: 30
Likes: 0
Received 0 Likes
on
0 Posts
email address for Bill Fowler
Anyone out there got a email addr. for bill Fowler, the TSB investigator.
Worked with him in the 90's - excellent operator and real gent
tks
Worked with him in the 90's - excellent operator and real gent
tks
Join Date: Sep 2001
Location: 38N
Posts: 356
Likes: 0
Received 0 Likes
on
0 Posts
aardvark2zz wrote: The on board computer, with a very simple software, could warn them of serious danger ahead as they are accelerating.
The informal discussion here has clearly pointed out that quite a number of factors can influence the ability of the bird to depart her runway in the time and space available. Weight, elevation, temperature, humidity, thrust, wheel friction, winds, turbulence, runway surface contamination, etcetera. A large number of variable but potentially decisive factors affect the launch outcome. Most are hard to know precisely and uncontrollable once the roll begins. The choices mid-roll devolve to power, airfoils, and reject/braking.
Departure performance is pre-calculated with an expected weight, thrust, slope, winds, temps, etc. The calculation, as done manually, creates a few numbers such as MTOW and V-speeds. But the same information and the same calculation can just as easily create a complete moment-by moment profile curve, expressed as speed vs distance or acceleration vs time or roll distance vs time from throttle up.
Having this computed curve of expected performance at hand, it is the most natural thing in the world for a little bit of electronics to compute a moment-by moment 'actual performance' during the roll from airspeed, surface radar, gps, wheel spin, acceleration, or all of the above. By continuously comparing the projected performance with the observed values, the resulting real-time all-encompassing magical performance number would provide a very true comparison of expected versus actual progress in safely unsticking from mother earth. And the information - at least for gross deviations in performance - would be available and useful early in the roll, making an informed take-off abort possible before the uncertainty of stopping makes scrubbing the takeoff a terribly hard decision.
This evolving launch data could be plotted in detail on the FE console, if present, but a few lights would probably serve the purpose up front. Green, amber, red and brown might be appropriate.
What this bit of kit would do is add method and precision to just what a pilot does now: monitor the progress of the takeoff run and determine whether it "feels" right or not. The difference is that it would add a degree of accountable precision to a seat-of-the-pants process and work just as well in cold dark places, in rain and snow and other cases where crew sensory perception is constrained.
In an era where good electronics can be created cheaper and easier than good plumbing, this is not an expensive thing to do.. in comparison to the cost of a tragically failed departure. Certainly it can be integrated into aircraft electronic systems, but that is a long-cycle evolution. For sooner availability, the necessary bits could be made fairly cheap and reliable as an add-on black box or even as a portable unit for application on all those ageing airframes hauling heavy stuff in the night.
The history of aviation is trial and error. The best monument to those lost in the process is to SOLVE the problems that tragedy reveals. RIP.
Last edited by arcniz; 27th Oct 2004 at 20:42.
Put time to V1 in the performance tables
Yes, I know that it will cost a forest full of trees to republish takeoff performance tables, but the computational task is straightforward.
Having the PNF monitoring the stopwatch up to V1 as well as airspeed and engine parameters can catch all of these nasty conditions during takeoff while there's still time to stop the beast on the remaining runway.
The stopwatch has to be monitored during non-precision approaches; so, it looks like time to begin doing it on takeoffs as well.
And yes, the glass cockpit jobs can integrate time to V1 into the FMC and flight display.
Having the PNF monitoring the stopwatch up to V1 as well as airspeed and engine parameters can catch all of these nasty conditions during takeoff while there's still time to stop the beast on the remaining runway.
The stopwatch has to be monitored during non-precision approaches; so, it looks like time to begin doing it on takeoffs as well.
And yes, the glass cockpit jobs can integrate time to V1 into the FMC and flight display.
Join Date: Jan 2001
Location: Dallas, TX USA
Posts: 739
Likes: 0
Received 0 Likes
on
0 Posts
Arcniz, here are some links to research into the concept of a Takeoff and Performance Monitoring System, which you discribe:
NASA TOPMS research
University of Saskatchewan thesis on TOPMS system
NASA TOPMS cockpit display development
As you can see, the concept of a Takeoff and Performance Monitoring System is being worked on by some groups. That's a good thing, because I think it's a great idea.
I also agree with RatherBeFlying, as long as the time vs speed parameters can be derived fairly easily, the stopwatch will work until a more sophisticated system can be created.
(edited to add a comment)
NASA TOPMS research
University of Saskatchewan thesis on TOPMS system
NASA TOPMS cockpit display development
As you can see, the concept of a Takeoff and Performance Monitoring System is being worked on by some groups. That's a good thing, because I think it's a great idea.
I also agree with RatherBeFlying, as long as the time vs speed parameters can be derived fairly easily, the stopwatch will work until a more sophisticated system can be created.
(edited to add a comment)
Last edited by Flight Safety; 28th Oct 2004 at 18:43.
None but a blockhead
Join Date: Nov 1999
Location: London, UK
Posts: 535
Likes: 0
Received 0 Likes
on
0 Posts
How accurately is engine thrust known? It could all boil down to something as simple as F=MA - if you know any two of those, you can instantly derive the third. If the aircraft knows that it's putting out so many kilograms of thrust and it's accelerating at so many metres per second squared, then it's effectively weighed itself - and will be able to spot that it's operating outside the safe envelope, or if the input weight to the FMC is at gross variance with the empirically derived figure. Much simpler than arcniz' proposal of monitoring multiple factors during take-off, but perhaps useful nonetheless - and automatic, which may be a bonus if you drop the stopwatch.
Couldn't such a system be achievable with some extra software in the existing FMCs and no hardware changes at all? Not being anything other than an armchair consumer of avionic information, I don't know how the calibration would work or whether there'd be any point in trying to derive the normal operating parameters of the aircraft by routinely monitoring performance and building up a database of observed readings, but the physics is at heart as simple as it gets.
Perhaps this might be a quick yet effective first step in catching this class of accident. I could even see it being a task that something as simple as a PDA with GPS could cope with (but that's not to understate the problem of false positive alarms). A full TOPMS system has to be a good idea, but sometimes you get a lot of the benefits of a complex idea with a relatively small core feature set.
R
Couldn't such a system be achievable with some extra software in the existing FMCs and no hardware changes at all? Not being anything other than an armchair consumer of avionic information, I don't know how the calibration would work or whether there'd be any point in trying to derive the normal operating parameters of the aircraft by routinely monitoring performance and building up a database of observed readings, but the physics is at heart as simple as it gets.
Perhaps this might be a quick yet effective first step in catching this class of accident. I could even see it being a task that something as simple as a PDA with GPS could cope with (but that's not to understate the problem of false positive alarms). A full TOPMS system has to be a good idea, but sometimes you get a lot of the benefits of a complex idea with a relatively small core feature set.
R
Some Dassault products already display accelaration in G on the PFD.
In these aircraft, the expected Accelaration is provided as part of the Takeoff Data computation, but not often used day to day.
Time to 100kts used to be a stopwatch check used in my dim dark past.
A difference between calculated and actual times was a pretty good reason for stopping. Don't see it considered much in the civilian world.
In these aircraft, the expected Accelaration is provided as part of the Takeoff Data computation, but not often used day to day.
Time to 100kts used to be a stopwatch check used in my dim dark past.
A difference between calculated and actual times was a pretty good reason for stopping. Don't see it considered much in the civilian world.
Join Date: Jan 2002
Location: France
Posts: 9
Likes: 0
Received 0 Likes
on
0 Posts
B747 Plane crash at Halifax NS - YHZ
I am not a pilot, just a crew member. I have followed this thread avidly because I have a strong connection with one of the people involved. These questions may appear superfluous to the experienced and for that I apologise.
There are two things that I don't understand.
When flying a 747-200 there are two pilots and a flight engineer who all have to agree on the EPR, is that right?
Secondly, if it is known that the aircraft is heavy and that at the end of the runway is a bern built out of concrete, can the EPR not be set so that the engines have full thrust regardless - or are there likely to be consequences of setting the engines to full power even if it may not be required.
There are two things that I don't understand.
When flying a 747-200 there are two pilots and a flight engineer who all have to agree on the EPR, is that right?
Secondly, if it is known that the aircraft is heavy and that at the end of the runway is a bern built out of concrete, can the EPR not be set so that the engines have full thrust regardless - or are there likely to be consequences of setting the engines to full power even if it may not be required.
Join Date: Dec 1999
Location: England
Posts: 144
Likes: 0
Received 0 Likes
on
0 Posts
Janeee,
The thrust is always calculated and set so that you will be airborne at the end of the runway by a safe margin.
Something here has gone wrong. Either the aircraft was heavier than the paperwork said, or something else has happened to the same effect.
It is sometimes only apparent to the pilots that something's wrong when it's too late. They may have applied full thrust as a last minute bid to save the day, but it didn't work.
Hope this helps,
Stu
The thrust is always calculated and set so that you will be airborne at the end of the runway by a safe margin.
Something here has gone wrong. Either the aircraft was heavier than the paperwork said, or something else has happened to the same effect.
It is sometimes only apparent to the pilots that something's wrong when it's too late. They may have applied full thrust as a last minute bid to save the day, but it didn't work.
Hope this helps,
Stu
Join Date: Jan 2001
Location: Dunstable, Beds UK
Posts: 545
Likes: 0
Received 0 Likes
on
0 Posts
Reading this thread one gets the impression that the professional load master leaps out with his set of scales. weight the boxes, secures them on the pallet, adds up the total pallet weight, and loads the pallets into the correct pallet postion on the aircraft to allow for floor weight and C. of G.
I must have been in the wrong operation
On a 1 hour turnaround or through stop the LM had 1 hour to get the pallets off and load the new ones
on (23/28 pallets).
All he has to work on is the declared pallet weight.
If thats out he is out !
I would be interested to see how long the aircraft and loadmaster spent in Halifax and how many pallets he loaded.
If there is an error it is not likely to be the LM, it will be underdeclared pallets weights
On some of our operations we loaded the pallets in our own facility but in some cases we just picked up from the forwarder/ handling agent
I must have been in the wrong operation
On a 1 hour turnaround or through stop the LM had 1 hour to get the pallets off and load the new ones
on (23/28 pallets).
All he has to work on is the declared pallet weight.
If thats out he is out !
I would be interested to see how long the aircraft and loadmaster spent in Halifax and how many pallets he loaded.
If there is an error it is not likely to be the LM, it will be underdeclared pallets weights
On some of our operations we loaded the pallets in our own facility but in some cases we just picked up from the forwarder/ handling agent
Join Date: Jul 2004
Location: Cloud 9
Posts: 2,948
Likes: 0
Received 0 Likes
on
0 Posts
Golden Lion,
Why on earth would anyone tech-stop a B747 on an 8,500ft runway when you have Bangor, Goose & Gander all available and with longer runways?
It is common knowledge that Halifax is the seafood pick-up and that's what they were doing there.
Why on earth would anyone tech-stop a B747 on an 8,500ft runway when you have Bangor, Goose & Gander all available and with longer runways?
It is common knowledge that Halifax is the seafood pick-up and that's what they were doing there.
Join Date: Oct 2004
Location: Asia
Posts: 12
Likes: 0
Received 0 Likes
on
0 Posts
CargoOne: “Any news on FDR/CVR readout?”
The Transportation Safety Board of Canada said the MK Airlines jet was rolling down the runway at 150 mph, about 35 mph slower than it should have been, when it crashed Oct. 14 at Halifax International Airport, the Globe and Mail reported Saturday.
"In order to get to (the minimum 184 mph) within the available runway, they needed more thrust," Bill Fowler of the TSBC said Friday. The jumbo jet was barely airborne when it rocketed off the end of the runway and crashed into a wooded area, killing all seven crew members.
He said a flight data recorder recovered from the wreckage last weekend indicated the 747's four engines were functioning normally. Fowler wouldn't speculate why the plane didn't have enough thrust.
The Transportation Safety Board of Canada said the MK Airlines jet was rolling down the runway at 150 mph, about 35 mph slower than it should have been, when it crashed Oct. 14 at Halifax International Airport, the Globe and Mail reported Saturday.
"In order to get to (the minimum 184 mph) within the available runway, they needed more thrust," Bill Fowler of the TSBC said Friday. The jumbo jet was barely airborne when it rocketed off the end of the runway and crashed into a wooded area, killing all seven crew members.
He said a flight data recorder recovered from the wreckage last weekend indicated the 747's four engines were functioning normally. Fowler wouldn't speculate why the plane didn't have enough thrust.
Join Date: Jul 2000
Location: U.S.A.
Posts: 474
Likes: 0
Received 0 Likes
on
0 Posts
While not jumping to any conclusions on the cause of this accident, 747Focal’s post is what I considered a high probability cause. It has been my experience that in three crewmember aircraft where the TO Thrust setting is manually computed by the FE, rarely is there a procedural or operational crosscheck of the values presented. Now in this situation (relatively short runway/heavy payload), EPR should have been near/at MAX allowable.
Time will tell all.
Time will tell all.
Join Date: Aug 2004
Location: Brisbane
Posts: 52
Likes: 0
Received 0 Likes
on
0 Posts
How difficult/expensive can it be to modify the onboard computer to calculate weight. The aircraft has an undercarriage, the undercarriage strut requires oil at pressure, so why not put three sensors onto the strut, link the sensors to the computer and the computer calculates the current PSI which is then converted to a weight?
The Reverend
Join Date: Oct 1999
Location: Sydney,NSW,Australia
Posts: 2,020
Likes: 0
Received 0 Likes
on
0 Posts
Queenslander, most 747 freighters have a weight and balance computer fitted. The system weighs the airplane, computes the centre of gravity of the airplane and displays the gross weight and the MAC on indicators on P4 Flight Engineer's panel. The indications are cross checked against the load sheet and have to meet certain tolerances before flight. There are 18 weight sensors located in the landing gear axels, one for each wheel and an attitude sensor which senses the longitudinal attitude angle of the airplane. My last nine years of aviation career was spent operating 747-200F aircraft.