SQ-368 (engine & wing on fire) final report out
Join Date: Dec 2000
Location: I used to know
Posts: 175
Likes: 0
Received 0 Likes
on
0 Posts
Touchdown was at 0649. The fire was observed approx 20 seconds later. The fire service arrived 57 seconds after the aircraft stopped. ( a very quick and fortuitous time). The fire was put out at 0653 some 4 minutes after initiation.
Some may say a very quick time however to put that in perspective during the Manchester fire the hull was breached betwean 5 and 20 seconds after stopping and the windows gave within a minute of stopping.
Approx 5 Passengers had died from the effect of the flames and heat (the rest from the effects of the toxic smoke) by, at the latest 1:30 minutes after stopping.
Compare this to the timeline of the Singapore incident and work out what they were chatting about at that point.
Some may say a very quick time however to put that in perspective during the Manchester fire the hull was breached betwean 5 and 20 seconds after stopping and the windows gave within a minute of stopping.
Approx 5 Passengers had died from the effect of the flames and heat (the rest from the effects of the toxic smoke) by, at the latest 1:30 minutes after stopping.
Compare this to the timeline of the Singapore incident and work out what they were chatting about at that point.
Join Date: Dec 2000
Location: I used to know
Posts: 175
Likes: 0
Received 0 Likes
on
0 Posts
The Captain knows he has a fire despite not having any fire indications in the cockpit, because the tower has told him so.
Apart from giving a "cabin crew at stations" call ( or the Singapore equivalent) what does the crew do?
Contact is established with the FC at 0651:50 around 2 minutes after being told of a fire. No attempt to contact the cabin and establish what they could see. No attempt to turn on the video. No reported attempt to ask the tower.
Instead wait for the FC who inspite of saying the fire is a big one recomends disembarkation not evacuation. The FC was confident he would get the fire under control, however fire is unpredictable. Just look at what happened to that poor firefighter who lost his life in Dubai.
The Captain who is initially evacuation minded, defers to the FC and effectively absolves himself of responsibility and allows the FC to become the decision maker. Perhaps he woild like to donate his salary to the FC as well.
Apart from giving a "cabin crew at stations" call ( or the Singapore equivalent) what does the crew do?
Contact is established with the FC at 0651:50 around 2 minutes after being told of a fire. No attempt to contact the cabin and establish what they could see. No attempt to turn on the video. No reported attempt to ask the tower.
Instead wait for the FC who inspite of saying the fire is a big one recomends disembarkation not evacuation. The FC was confident he would get the fire under control, however fire is unpredictable. Just look at what happened to that poor firefighter who lost his life in Dubai.
The Captain who is initially evacuation minded, defers to the FC and effectively absolves himself of responsibility and allows the FC to become the decision maker. Perhaps he woild like to donate his salary to the FC as well.
Join Date: Jul 2002
Location: Confusio Helvetica
Posts: 311
Likes: 0
Received 0 Likes
on
0 Posts
As for "shutting down an unhealthy engine"; that is discussed in the report and the answer seems to be that there was only 1 indication that was anything wrong, a low oil quantity alert. Other indicators showed it may be a defective indication as temperatures and oil pressures were OK.
And it's not true that the crew offloaded all their decisions to the ground:
At 0328 hrs, the engineer sent a message via the Aircraft Communication and Reporting System (ACARS) to the flight crew informing them about the recommendation by the technical services personnel for the aircraft to return to Singapore and requesting them to contact the engineering
control centre.
control centre.
Only on a second call do they switch the pack off.
They repeatedly told ATC that they had no problem, the FUEL DISAGREE message they disagreed with, and, when the FC told them to disembark from the left side of their burning aircraft, they wanted to make absolutely sure that they weren't supposed to evacuate. Heck, they didn't even use the bottles until ARFF asked them about it. Probably didn't think they were necessary.
I'm just sayin'. That's not offloading decision-making.
My favorite recommendation:
The aircraft manufacturer evaluate the need for providing guidance on
how to perform fuel leak check with the engines operated at unequal
thrust. [TSIB Recommendation RA-2017-023]
how to perform fuel leak check with the engines operated at unequal
thrust. [TSIB Recommendation RA-2017-023]
My favorite recommendation:
Quote:
The aircraft manufacturer evaluate the need for providing guidance on
how to perform fuel leak check with the engines operated at unequal
thrust. [TSIB Recommendation RA-2017-023]
Quote:
The aircraft manufacturer evaluate the need for providing guidance on
how to perform fuel leak check with the engines operated at unequal
thrust. [TSIB Recommendation RA-2017-023]
Join Date: Aug 2007
Location: Standby, Resyncing other FMC...
Posts: 131
Likes: 0
Received 0 Likes
on
0 Posts
For me the big question is:
If you have operated an engine in idle for a long time before landing due to suspected damage, why would you use the reverser on that engine?
If they hadn't used the reverser this thread would never happened.
If you have operated an engine in idle for a long time before landing due to suspected damage, why would you use the reverser on that engine?
If they hadn't used the reverser this thread would never happened.
Join Date: Jul 2002
Location: Confusio Helvetica
Posts: 311
Likes: 0
Received 0 Likes
on
0 Posts
tdracer, maybe I'm misunderstanding something, since it sounds like what you are describing is the Calculated fuel that, in fact, did catch the leak and did send an EICAS message, but was ignored.
I find it interesting to work through the logic. Here's what the report says:
And the footnote:
So, as I understand it, the totalizer and the calculated (via FQIS) numbers disagreed, and it kicked a message. The flight crew disregarded the message, because they assumed that the Calculated quantity was the FMS' version of the planned fuel remaining quantity, based on predicted performance according to what planned engine settings were and the aircraft's route. (Does the FMC use Calculated fuel as the basis for its Arrival Fuel Prediction?)
Since the FMS did not have an option for calculating performance based on one engine running at idle, they ran the numbers with one engine INOP.
So they figured that since
A) earlier, they found themselves with 600 kg more fuel than planned
B) the FMS was calculating performance for a route they were not taking, and
C) the right engine was consuming fuel at idle, rather than being shut down, as the FMS thought
that the FUEL DISAGREE message was spurious. They did some back-of-the-envelope calculations and decided they didn't need to do a fuel leak check, presumably because that would involve having to shut their engine down.
In other words, when they got the message, they made up a definition of "CALCULATED" fuel where the calculations were based on the FMS instead of the EQIS. In their imaginary world, there wasn't a problem.
And that's why I'm amused by the recommendation. Someone no doubt who is not as ignorant as I will gladly explain to me:
1. Outside of ETOPS certification, why on a twin with one good engine and near several suitable airports, would it make more sense to reduce a malfunctioning engine to idle than to shut it down?
2. What the benefit is of a fuel leak test with both engines operating, but asymmetrically? How often does a FUEL DISAGREE EICAS message pop up on an already-malfunctioning engine without there being a fuel leak?
In the report's defense, it does make the comment:
I find it interesting to work through the logic. Here's what the report says:
At 0521 hrs, the flight crew received a FUEL DISAGREE message on the EICAS. The flight crew performed the FUEL DISAGREE checklist. The FUEL DISAGREE checklist suggested four scenarios in which a fuel leak should be suspected and when the flight crew should perform the FUEL LEAK checklist. One such scenario is when the TOTALIZER fuel quantity is less than the CALCULATED.
The CALCULATED fuel quantity is determined by the flight management computer by subtracting the fuel used (calculated basing on fuel flow figures as measured by sensors in the engines) from the total fuel quantity at the start of the flight fuel quantity.
Since the FMS did not have an option for calculating performance based on one engine running at idle, they ran the numbers with one engine INOP.
So they figured that since
A) earlier, they found themselves with 600 kg more fuel than planned
B) the FMS was calculating performance for a route they were not taking, and
C) the right engine was consuming fuel at idle, rather than being shut down, as the FMS thought
that the FUEL DISAGREE message was spurious. They did some back-of-the-envelope calculations and decided they didn't need to do a fuel leak check, presumably because that would involve having to shut their engine down.
In other words, when they got the message, they made up a definition of "CALCULATED" fuel where the calculations were based on the FMS instead of the EQIS. In their imaginary world, there wasn't a problem.
And that's why I'm amused by the recommendation. Someone no doubt who is not as ignorant as I will gladly explain to me:
1. Outside of ETOPS certification, why on a twin with one good engine and near several suitable airports, would it make more sense to reduce a malfunctioning engine to idle than to shut it down?
2. What the benefit is of a fuel leak test with both engines operating, but asymmetrically? How often does a FUEL DISAGREE EICAS message pop up on an already-malfunctioning engine without there being a fuel leak?
In the report's defense, it does make the comment:
During the initial training to operate this aircraft, the operator provides training to all its pilots to understand the requirements of the FUEL
DISAGREE checklist. However, in this case, the flight crew appeared to have misinterpreted certain requirements of this checklist even though they have undergone the training.
DISAGREE checklist. However, in this case, the flight crew appeared to have misinterpreted certain requirements of this checklist even though they have undergone the training.
Join Date: Dec 2000
Location: I used to know
Posts: 175
Likes: 0
Received 0 Likes
on
0 Posts
Presumably if they had shut down the engine they would have needed to follow the checklist and land at the nearest suitable airport which would not have been Singapore. The recommendations from base were no doubt influenced by the same desire to get the aircraft back to base.
Join Date: Jan 2008
Location: Netherlands
Age: 46
Posts: 343
Likes: 0
Received 0 Likes
on
0 Posts
Engine failure lesson 1: An engine at idle still produces some electricy, hydraulics, bleed air and less drag. It is always better to have it at idle than to shut it down. You only shut it down if you really, really have to.
Didn't fly big machinery, but a tale of small twin turbine. In cruise and oil pressure dropped into the yellow and temp reduction of 10°. Reduced power to idle and pressure was in the bottom of allowable. No APU, kept engine running so as to have auxiliaries available, and should our "good" engine decide to take a holiday. Upon landing and shutting down ramp advised of a puddle of oil on the tarmac. Checking the dipstick no oil was evident. Cause, a split in the reservoir to pump line. So while the engine was running it was sucking in air and aerating the oil, explaining the drop in both pressure and temp. The instant of shutting down all the oil drained out of the split.
Earlier in life had been given the advice, "never shut down an engine capable of providing power".
Earlier in life had been given the advice, "never shut down an engine capable of providing power".
Dinger
The fuel disagree/leak checklists for the 777 are quite explicit that "calculated" for fault finding/QRH purposes is the instantaneous calculated on board fuel ( i.e. initial onload minus the integrated flow), and that is the number to be compared with the FQIS tanks totaliser figure; both are displayed for easy comparison, right next to each other, on FMS page Progress 2 (in our case). Since they are instantaneous, not predicted figures, there's no need for any future performance input/knowledge of route, etc.
On most flights the two figures will be well within a within a tonne of each other. Now from the report it seems the SIN crew had a 4 tonne mismatch (!!!!) so unfortunately it looks like they well and truely managed to over analyse things and talked themselves out of actioning the Fuel Leak checklist...however...
Don't want to offend tdracer or Boeing but you mention the comment in the report about the training of the fuel disagree/fuel leak checklist, and I'd say that could well be a fair cop.......FWIW for the 777 this has been a very "hot" item for trainers for a while at at least one outfit I know of. The checklists are somewhat protracted (for good reason), convoluted, but there are one or two interestingly worded questions and as a result using the electronic checklist it is quite easy to end up ticking the wrong box and going up a blind or misleading alley, or not going up an alley at all (such as going into the Fuel imbalance Checklist but then not going on to action the Fuel Leak Checklist....when in fact taking a step back it might be obvious it is the sensible thing to do).....
FWIW Arrival fuel prediction (FMS progress page 1) does use the calculated fuel minus predicted burn, winds, temps etc etc. There is an option, rarely used to switch to using the FQIS quantity..but in any event you shouldn't be using that predicted figure arrival as a decider for Fuel leak analysis, it's the instantaneous calculated fuel vs. gauged fuel tanks that matters..
The flight crew disregarded the message, because they assumed that the Calculated quantity was the FMS' version of the planned fuel remaining quantity, based on predicted performance according to what planned engine settings were and the aircraft's route.)
The fuel disagree/leak checklists for the 777 are quite explicit that "calculated" for fault finding/QRH purposes is the instantaneous calculated on board fuel ( i.e. initial onload minus the integrated flow), and that is the number to be compared with the FQIS tanks totaliser figure; both are displayed for easy comparison, right next to each other, on FMS page Progress 2 (in our case). Since they are instantaneous, not predicted figures, there's no need for any future performance input/knowledge of route, etc.
On most flights the two figures will be well within a within a tonne of each other. Now from the report it seems the SIN crew had a 4 tonne mismatch (!!!!) so unfortunately it looks like they well and truely managed to over analyse things and talked themselves out of actioning the Fuel Leak checklist...however...
Don't want to offend tdracer or Boeing but you mention the comment in the report about the training of the fuel disagree/fuel leak checklist, and I'd say that could well be a fair cop.......FWIW for the 777 this has been a very "hot" item for trainers for a while at at least one outfit I know of. The checklists are somewhat protracted (for good reason), convoluted, but there are one or two interestingly worded questions and as a result using the electronic checklist it is quite easy to end up ticking the wrong box and going up a blind or misleading alley, or not going up an alley at all (such as going into the Fuel imbalance Checklist but then not going on to action the Fuel Leak Checklist....when in fact taking a step back it might be obvious it is the sensible thing to do).....
Does the FMC use Calculated fuel as the basis for its Arrival Fuel Prediction?)
Last edited by wiggy; 18th Mar 2017 at 12:26.
Join Date: Apr 2008
Location: London
Posts: 24
Likes: 0
Received 0 Likes
on
0 Posts
@RCSA - it is simple. The risks of fire are so unpredictable and great, and the rapidity at which toxic effects can spread, is that the decision not to evacuate even with this 'safe' outcome is far worse than evacuating immediately before the option is taken away. BA in Vegas is a sterling example of how to do it properly. Singapore were lucky. Aviation does not need lucky.
Nemo Me Impune Lacessit
Join Date: Jun 2004
Location: Derbyshire, England.
Posts: 4,091
Received 0 Likes
on
0 Posts
Absolutely no similarities between the BA incident in Las Vegas and the SQ incident at Changi - the evacuation in Las Vegas wasn't so hot either, pax walking away with their carry on luggage, imagine the congestion in the aisles that must have caused.
You would think that after the Sewol capsize, where pax were instructed to remain their cabins and were later found drowned, that captains would be more inclined to evacuate before it's too late.
The lesson has not been learned in Singapore.
Next time the luck might not hold.
The lesson has not been learned in Singapore.
Next time the luck might not hold.
Last edited by RatherBeFlying; 19th Mar 2017 at 05:01.
Join Date: Oct 2001
Location: Lost, but often Indonesia
Posts: 652
Received 0 Likes
on
0 Posts
Would a windmilling engine at cruise (i.e. shut down) still produce some degree of hydraulic pressure and electrical generation? What sort of speed would the engine be rotating at compared to idle? Or does a complete shutdown isolate everything? Just trying to understand what is to be gained by leaving a sick engine running at idle for hours. Apologies if it's a daft question...
On a general point one consideration certainly on a twin is if the "sick" engine is still capable in extremis of producing useable thrust if needed then it might be sensible to leave it running......as long as procedures don't say otherwise.
However in the context of this incident they had more than a slightly "sick" engine...certainly as I read the report if the crew had gone through the Fuel Disagree checklist fully and correctly they would have been led onwards into actioning the Fuel Leak checklist, at which point they hopefully would have realised that the engine was more than just slightly "sick". The QRH would have been telling the crew they had an engine Fuel Leak and that they needed to shut it down, not leave it at idle...but for reasons various they never got as far as the Fuel Leak checklist (see my previous comment about the 777 checklist).
Last edited by wiggy; 21st Mar 2017 at 11:36.
TBH I'd be sticking my neck out offering up numbers since I haven't seen a real windmilling engine in a long long time 
and never on a real 777. Might be one for those who do or who involved in airtests to answer.
and never on a real 777. Might be one for those who do or who involved in airtests to answer.
Last edited by wiggy; 21st Mar 2017 at 11:35.
