AF 777 wrong weight inputs, off by 100 tonnes?
Join Date: Nov 2009
Location: flying by night
Posts: 500
Likes: 0
Received 0 Likes
on
0 Posts
Sensors capable of measuring pressure on each landing gear to calculate weight exist, have been installed and tested. Ground based systems have been tested as well. Would be nice to have as an extra layer of safety; that we don't see a lot of them is about cost I guess. Cf post #10
Load weigh systems.
Earthmoving trucks have had loadweigh systems for quite a while.
Pressure transducers in the struts and a control (summing) module - not such a great expense.
Not sure that accuracy would be suitable for more than a cross-check, but the hard part for the trucks is the slope and condition of the ground that they're sitting on.
That may not be a problem for aircraft on the apron. Or I hope not.
Pressure transducers in the struts and a control (summing) module - not such a great expense.
Not sure that accuracy would be suitable for more than a cross-check, but the hard part for the trucks is the slope and condition of the ground that they're sitting on.
That may not be a problem for aircraft on the apron. Or I hope not.
In my airline the calculations are made completely independently by the F/O and the Captain, on seperate Performace tools, with no discussion. Each individual sources the zero fuel weight figure independently, adds the fuel amount, adds a buffer for last minute changes and then at a time that suits the crew say " what did you get?" And a comparison is made. Once in a while the figures are a bit different because someone has not accounted for a wet runway or something similar and further probing ensues in order to determine who is correct. The system works well and the greatest weakness in it lies in the load control office where the zero fuel weight is calculated.
The second greatest weakness is in an individual's personal discipline, it is tempting to ask the gross weight used by the f/o, this should not be done.
I had assumed that after EK 407 that this was how most airlines did things. I would be very interested to know if I am wrong.
Interruptions and chatter on the flight deck during set up is also a threat and ground staff/ cabin members need to be disciplined in this area.
The second greatest weakness is in an individual's personal discipline, it is tempting to ask the gross weight used by the f/o, this should not be done.
I had assumed that after EK 407 that this was how most airlines did things. I would be very interested to know if I am wrong.
Interruptions and chatter on the flight deck during set up is also a threat and ground staff/ cabin members need to be disciplined in this area.
Join Date: Apr 2010
Location: London
Posts: 7,072
Likes: 0
Received 0 Likes
on
0 Posts
KENV wrote
"A system with that level of precision is really easy to design and build.
A system with that level of accuracy under all loading conditions over several years of service? Not so much."
You are correct - but I think that someone like Mr Dyson could come up with something new and accurate pretty quickly - it isn't rocket science
The problem is that no-one sees it as a real prioirity
"A system with that level of precision is really easy to design and build.
A system with that level of accuracy under all loading conditions over several years of service? Not so much."
You are correct - but I think that someone like Mr Dyson could come up with something new and accurate pretty quickly - it isn't rocket science
The problem is that no-one sees it as a real prioirity
Accuracy ?
KENV : Nothing that occasional re-calibration wouldn't fix.
But I would still see it only as a double check (treble check?) against existing procedures
But I would still see it only as a double check (treble check?) against existing procedures
Rather than adding hardware, couldn't the existing computers be programmed to compare the aircraft's actual acceleration to that predicted? Surely the existing sensors are accurate enough to permit detection of a major weight error early in the takeoff roll.
But as SLF I have no idea whether this would be solving a real problem, or whether additional automation would just exacerbate the underling problem (if there is one).
But as SLF I have no idea whether this would be solving a real problem, or whether additional automation would just exacerbate the underling problem (if there is one).
Join Date: Jun 2008
Location: 31000FR
Posts: 60
Likes: 0
Received 0 Likes
on
0 Posts
Interruptions and chatter on the flight deck during set up is also a threat and ground staff/ cabin members need to be disciplined in this area.
Here you have the root cause: The last 5 minutes before STD/push back are the most stressed phase of flying, and the load figures often come at the last minute.
In most of "my" airlines I have seen how the stress makes the guys load-shed during the calculations. Why? Because there isn't enough time if you want to leave "on time", and OTP is the big mantra in most airlines.
In one airline we came up with a smart system: If you entered a GW less than MZFW it would block and give a warning. One day the crew attempted to enter the ZFW iso TOW, and when the warning came up they simply did an override. Result: Very close to disaster and a big hole in the tail. This is what pre-departure stress can do to you.
The OBAWS (Onboard Balance & Weight System) has been around for ages and is an excellent x-check reference, but the "Safety-is-our-main-concern" managements are very reluctant to pay the several thousand dollars for a system that does not bring revenue and whose sole purpose is to catch mstakes made by the excessively high paid pilots.
The last line of defence is the Speed Trend Vector found in most modern aircraft. If you don't accelerate 4KTS/sec you are unlikely to become airborne in 3000m, so you either go TOGA or abort.
During SIM trials in a widebody quad I gave the crews wrong figures which only gave them 60% of the required acceleration: They all overran the runway....In the 2nd attempts (and after a briefing) they all survived when going TOGA when they identified the slow acceleration (around 80-100 kts) - and they did not scrape the tail (but some close calls).
PS: The frog discussion is rubbish. I have checked several AF crews and they perform better than the average, but as with most of the old legacy carriers complacency is hard to eradicate. Not a unique AF issue.
Here you have the root cause: The last 5 minutes before STD/push back are the most stressed phase of flying, and the load figures often come at the last minute.
In most of "my" airlines I have seen how the stress makes the guys load-shed during the calculations. Why? Because there isn't enough time if you want to leave "on time", and OTP is the big mantra in most airlines.
In one airline we came up with a smart system: If you entered a GW less than MZFW it would block and give a warning. One day the crew attempted to enter the ZFW iso TOW, and when the warning came up they simply did an override. Result: Very close to disaster and a big hole in the tail. This is what pre-departure stress can do to you.
The OBAWS (Onboard Balance & Weight System) has been around for ages and is an excellent x-check reference, but the "Safety-is-our-main-concern" managements are very reluctant to pay the several thousand dollars for a system that does not bring revenue and whose sole purpose is to catch mstakes made by the excessively high paid pilots.
The last line of defence is the Speed Trend Vector found in most modern aircraft. If you don't accelerate 4KTS/sec you are unlikely to become airborne in 3000m, so you either go TOGA or abort.
During SIM trials in a widebody quad I gave the crews wrong figures which only gave them 60% of the required acceleration: They all overran the runway....In the 2nd attempts (and after a briefing) they all survived when going TOGA when they identified the slow acceleration (around 80-100 kts) - and they did not scrape the tail (but some close calls).
PS: The frog discussion is rubbish. I have checked several AF crews and they perform better than the average, but as with most of the old legacy carriers complacency is hard to eradicate. Not a unique AF issue.
The OBAWS (Onboard Balance & Weight System) has been around for ages
I believe that the system on the -400F was better and more reliable.
Join Date: Aug 2002
Location: Up north
Posts: 1,657
Likes: 0
Received 0 Likes
on
0 Posts
BTW, answer to post 26: a LOSA audit is being done as we speak, the second one in 3 years. And for the halfwits tinking like post 27, the B777/744 fleet manager offerded his resignation yesterday.
This is being taken seriously.
This is being taken seriously.
Nice attitude by the way, AF flight crew?
One would like to think that the fancy new load and thrust calculation programs could produce a time to V1.
Join Date: Mar 2007
Location: Europe
Posts: 256
Likes: 0
Received 0 Likes
on
0 Posts
Quoting ATSB report about Emirates weight/speeds errors on Melbourne take off (100 t error):
Incorrect weight 262.9 t : V1=143 VR=145 V2 = 154 FLEX temp = 74°
Actual weight 362.9 t : V1=149 VR = 161 V2=173 FLEX temp = 43°
In previous 2 months, pilots used 3 A340 types, weight ranging from 150 to 370 t and quoting one pilot, numbers are loosing their meaning, they become only numbers...
Still bingo ?
Incorrect weight 262.9 t : V1=143 VR=145 V2 = 154 FLEX temp = 74°
Actual weight 362.9 t : V1=149 VR = 161 V2=173 FLEX temp = 43°
In previous 2 months, pilots used 3 A340 types, weight ranging from 150 to 370 t and quoting one pilot, numbers are loosing their meaning, they become only numbers...
Still bingo ?
Shouldn't 74° temperature have been a clue?
Join Date: Feb 2008
Location: Stockport
Age: 84
Posts: 282
Likes: 0
Received 0 Likes
on
0 Posts
One would like to think that the fancy new load and thrust calculation programs could produce a time to V1.
Join Date: Sep 2011
Location: earth
Posts: 1,098
Likes: 0
Received 0 Likes
on
0 Posts
The 77L in its freight version has some very special characteristics that make any 'gut feeling' or 'V-speed vs. weight dead reckoning' difficult.
As it is equipped with the 110lbs GE90 engines but the short body and a very light DOW, the Vmc on ground and in the air is extremely delicate. Imagine an empty flight TOW (145 tons absolutely possible vs. max TOW 347t) one can imagine what an engine failure and subsequent max power on one of these power plants means aerodynamically.
Therefore the Vmin's for such take-offs and landings requires speeds in excess of up to 20kts more than what the weight might normally dictate. The same effect happens with the assumed temperature, which remains fairly low in such cases.
This is not to excuse the crew, but it makes error detection with techniques mentioned in many contributions above quite tricky.
Cross qualified flying with animals so different in weight and characteristics, as well as a huge spread of cockpit equipment will always lead to errors. The Airbus 32/33/34 family has a long history and now the T7 family joins with ever more versions of airframes, engines and cockpit equipment. The trapping of errors is that much more demanding. Including more calls or time checks during take-off, or a quadruple crosschecking during calculations only increase workload and shed the initial shortcomings like short cuts in training, lack of basic experience and most urgently fatiguing rosters.
As it is equipped with the 110lbs GE90 engines but the short body and a very light DOW, the Vmc on ground and in the air is extremely delicate. Imagine an empty flight TOW (145 tons absolutely possible vs. max TOW 347t) one can imagine what an engine failure and subsequent max power on one of these power plants means aerodynamically.
Therefore the Vmin's for such take-offs and landings requires speeds in excess of up to 20kts more than what the weight might normally dictate. The same effect happens with the assumed temperature, which remains fairly low in such cases.
This is not to excuse the crew, but it makes error detection with techniques mentioned in many contributions above quite tricky.
Cross qualified flying with animals so different in weight and characteristics, as well as a huge spread of cockpit equipment will always lead to errors. The Airbus 32/33/34 family has a long history and now the T7 family joins with ever more versions of airframes, engines and cockpit equipment. The trapping of errors is that much more demanding. Including more calls or time checks during take-off, or a quadruple crosschecking during calculations only increase workload and shed the initial shortcomings like short cuts in training, lack of basic experience and most urgently fatiguing rosters.
Last edited by glofish; 8th Jun 2015 at 15:16.
Join Date: Aug 2014
Location: New Braunfels, TX
Age: 70
Posts: 1,954
Likes: 0
Received 0 Likes
on
0 Posts
KENV : Nothing that occasional re-calibration wouldn't fix.
Then look at it from a manager's point of view. What does this system do, REALLY? It backs up the pilot/FE to a make sure they made no gross errors. That's a lot of expense to check up on pilots that cost the airline a lot of money. Putting in a good procedural cross check (rather than a hardware cross check) is likely more effective AND cheaper.
Join Date: Aug 2014
Location: New Braunfels, TX
Age: 70
Posts: 1,954
Likes: 0
Received 0 Likes
on
0 Posts
Earthmoving trucks have had loadweigh systems for quite a while.
Pressure transducers in the struts and a control (summing) module - not such a great expense.
Pressure transducers in the struts and a control (summing) module - not such a great expense.
And more than just weight is involved. For the airplane, CG is involved. Many of the same weight numbers drive the CG calculation. And unlike the truck, improper CG can kill people on an airplane. (Witness NA Flt 102)
And I also said .........
KenV :
Please don't cherry pick.
I also said "But I would still see it only as a double check (treble check?) against existing procedures".
I did not suggest at any time that it could be used as a replacement for proper calculation.
At least until more sophisticated systems are developed.
However these systems are actually quite accurate and reliable.
Like everything else, they need maintenance and checking (or as I flippantly said, occasional re-calibration).
But more to the point, if your manual system is so bl...dy foolproof then why has there been two or more recent cases of such major, and potentially fatal, errors.
Even a poorly calibrated loadweigh would pick a 100 tonne finger fault.
Please don't cherry pick.
I also said "But I would still see it only as a double check (treble check?) against existing procedures".
I did not suggest at any time that it could be used as a replacement for proper calculation.
At least until more sophisticated systems are developed.
However these systems are actually quite accurate and reliable.
Like everything else, they need maintenance and checking (or as I flippantly said, occasional re-calibration).
But more to the point, if your manual system is so bl...dy foolproof then why has there been two or more recent cases of such major, and potentially fatal, errors.
Even a poorly calibrated loadweigh would pick a 100 tonne finger fault.
Join Date: Apr 1999
Location: Manchester, UK
Posts: 1,958
Likes: 0
Received 0 Likes
on
0 Posts
Capt Turbo's post hit the nail. These calculations nearly always have to be done in the five minutes before departure as that's when the load Figures come, and in some airlines the performance computer is also the electronic tech log, which is grasped in the engineer's oily hand until then. There are usually dispatchers, cabin crew, push-back crew, engineers, "on-time managers" (really!) and ATC all competing for attention so the flight deck is anything but a place of quiet concentration.
Join Date: Aug 2014
Location: New Braunfels, TX
Age: 70
Posts: 1,954
Likes: 0
Received 0 Likes
on
0 Posts
KenV :Please don't cherry pick.
I did not suggest at any time that it could be used as a replacement for proper calculation.
However these systems are actually quite accurate and reliable.
But more to the point, if your manual system is so bl...dy foolproof then why has there been two or more recent cases of such major, and potentially fatal, errors.
Even a poorly calibrated loadweigh would pick a 100 tonne finger fault.