Air Florida 90 DCA (continued from previous closed thread)

[andrasz]

Actually, the Air Florida crash was caused by icing on the wings which led to a stall right after takeoff.

Ice/snow contamination was just a contributing cause. The crew failed to use engine anti-ice during taxi and takeoff, and the slush-clogged engine probes provided erroneous EPR. A reduced thrust takeoff was done per noise abatement procedures with throttles set by the false high EPR reading, and the crew did not realise when the aircraft failed to climb that they had another 30% thrust available. Had they applied max power, the aircraft would have made it.

The similarity is that in both cases the crew executed a reduced thrust takeoff, and when the aircraft did not perform as expected, failed to apply available full power. Fortunately for the China Airlines crew, in their case there was no bridge to clear...

(suggest you move this to the end of the previous thread)

[Spad13]

Had they applied max power, the aircraft would have made it.

I would add the word "probably" before the "would have" part. I say that because it's difficult to know just how contaminated the aircraft was.

[andrasz]

I would add the word "probably" before the "would have" part. I say that because it's difficult to know just how contaminated the aircraft was.

Concur, though if I'm correct all post-accident simulator trials resulted in a successful takeoff if full power was applied at stickshaker, regardless of the degree of contamination. It was snowing, but temperature was above freezing (which is why crew did not go back to de-ice after 1:45 holding).

[3WE]

I would add the word "probably" before the "would have" part. I say that because it's difficult to know just how contaminated the aircraft was.

It was more than "probably"! I guess you can't say 100.00000%, BUT:

-The aircraft was in a long line of others, with everyone before them executing reasonably normal, highly-succesful departures and I don't recall anything being said to suggest that all these other aircraft deiced while AF did not.

Then, add to that that:

-all the flight data suggested that the plane was flying underpowered
-all the engine data suggested that the plane was flying underpowered

Sort of seems like engine power is what made AF different from all the other aircraft, not snow on the surfaces/engines.

Also, there's a ton of swiss cheese nuances in this crash:

The crew had one of the deicing options (engine anti-ice?) turned off when normal procedures said it should be on. That is what lead to the iced-over probe that lead to the incorrect power settings.

I also recall that another aircraft was on "short final", which might have lead to an unwillingness to abort the takeoff when it was initially noticed as being a little "cold" or underpowered.

Finally, this takeoff & crash had lots of CRM failures. The pilots clearly knew something was wrong- ("it's cold, real cold" referring to sluggish acceleration). However, the crew continued the takeoff, failed to interpret the engine instruments, and failed a final "no brainer" that to hell with what any instrument says- if you are failing at climbing and about to crash- firewall the throttles.

All that being said- it is a tough scenario when you swear you've done everything right, but the plane is acting in a way you've never seen it act. Sadly, that leads to thnking we can make it and and both pilots doing an "Oh-shit" tunnel vision riviting of their attention on the altimiter and/or bridge you are about to smack, as opposed to checking the instruments and remembering to shove the throttles full forward.

[Spad13]

It was more than "probably"! I guess you can't say 100.00000%, BUT:

-The aircraft was in a long line of others, with everyone before them executing reasonably normal, highly-succesful departures and I don't recall anything being said to suggest that all these other aircraft deiced while AF did not.

Then, add to that that:

-all the flight data suggested that the plane was flying underpowered
-all the engine data suggested that the plane was flying underpowered

Sort of seems like engine power is what made AF different from all the other aircraft, not snow on the surfaces/engines.

Sort of seems like, yes, but I bet the ice didn't help.

[Dmmoore]

The engines were not producing enough power to insure a successful takeoff on a dry day all other parameters being equal.
Was ice a factor? Yes. But not ice on the aircraft. Ice on the engine Pt2 probes.

[Foxtrot]

Another point I recall being mentioned was that the crew followed a DC 9 ahead of them too closely, which melted the snow on the leading edges and engne inlet, but it actually just refroze back on those surfaces during the takeoff roll.

[3WE]

Another point I recall being mentioned was that the crew followed a DC 9 ahead of them too closely, which melted the snow on the leading edges and engne inlet, but it actually just refroze back on those surfaces during the takeoff roll.

You've got that wrong. The crew was discussing the snow accumulation on the wing and made a passing comment that maybe he exhaust from the DC-9 ahead would clean thier wings.

It was just talk. It did not actually occur.

[Foxtrot]

You've got that wrong. The crew was discussing the snow accumulation on the wing and made a passing comment that maybe he exhaust from the DC-9 ahead would clean thier wings.

It was just talk. It did not actually occur.

This comes straight from the NTSB's summary on the accident:

WHILE WAITING, THE ACFT WAS PSND NEAR THE EXHAUST OF THE ACFT AHEAD.

And then Aviation Safety has it:

Although contrary to flight manual guidance, the crew attempted to deice the aircraft by intentionally positioning the aircraft near the exhaust of the aircraft ahead in line (a New York Air DC-9). This may have contributed to the adherence of ice on the wing leading edges and to the blocking of the engine’s Pt2 probes.

From the Washington post, which again has the same consistent statement:

http://www.washingtonpost.com/wp-dyn/content/article/2007/01/11/AR2007011102220.html

The NTSB found errors in the way the way the plane was de-iced -- the crew even tried to reduce the build-up on their Boeing 737 by using the exhaust of a jet in front of them.

Finally this from http://www.airdisaster.com/special/special-af90.shtml

'Palm 90' then proceeded to taxi into position behind a New York
Air DC-9, the last of sixteen aircraft in line for takeoff. With a light snow still falling, Wheaton decided to use the hot exhaust
from the preceeding DC-9's engines to melt the snow off the wings

I grant you the latter part of that statement (which I have excluded from posting here) is probably wrong based on further reports I've read on the accident, but it's basically stating what the other three reports/articles have said. Are all of those reports wrong about that? If they are I stand corrected.

[Leftseat86]

To this day I fail to see what logic prompted them to neglect turning the heat on.

[3WE]

To this day I fail to see what logic prompted them to neglect turning the heat on.

Not sure it was a "concious, delibarate and logical" choice that was made.

According to a TV documentary- (and maybe some CVR recordings...maybe), during one of the checklists, the engine anti-ice came up and the response was some sort of mumbling that sounded like "onuff")

Pure speculation, but maybe the other pilot heard "on"- and was assured it was on.

The only other explanation I remember being offered was that maybe the AF crews flew very little in snow, thus pilot or pilots might have fallen victim to doing what they "always do" takeoff after takeoff after takeoff.....anti ice....off!

[cegro27]

The crew became "complacent", a word that was used a lot to describe accidents involving pilot-error. They were doing something a certain way for so long (without any problems), that once something happens to change that normal routine, they become oblivious to any possible dangers that may arise from making a mistake. Like the NWA and DAL crashes in Detroit and Dallas respectively, where the pilots failed to "actually" check that the flaps were set for takeoff, "even though" they went over that item in the checklist before takeoff.

[Leftseat86]

Not sure it was a "concious, delibarate and logical" choice that was made.

According to a TV documentary- (and maybe some CVR recordings...maybe), during one of the checklists, the engine anti-ice came up and the response was some sort of mumbling that sounded like "onuff")

Pure speculation, but maybe the other pilot heard "on"- and was assured it was on.

The only other explanation I remember being offered was that maybe the AF crews flew very little in snow, thus pilot or pilots might have fallen victim to doing what they "always do" takeoff after takeoff after takeoff.....anti ice....off!

Yeah I've read the full reports and the excellent analysis found in Air Disaster Vol. 2 by Macarthur Job.

I just don't see the logic in skimming over the icing countermeasures when there's that much snow outside the window. It would seem obvious to me to pay extra attention to it.

[APS]

Correct me if I'm wrong, but wasn't there also the possibility that using Reverse-thrust to assist in pushing back from the gate might have rendered the de-icing applied to the aircraft ineffective/less effective?

[mzk49o1]

Correct me if I'm wrong, but wasn't there also the possibility that using Reverse-thrust to assist in pushing back from the gate might have rendered the de-icing applied to the aircraft ineffective/less effective?

I'm pretty sure, as it was mentioned that the reverse thrust would throw slush onto the wings.