Colgan Crash Experience

[n204wt]

Here's what Fox news is saying: "Colgan Air First Officer Rebecca Shaw, who has been with the airline since January 16, 2008, and had flown 2,244 hours with Colgan."
For the sake of discussion, let's say 16-JAN-08 is 392 days ago (365 days in a year plus 27 more for the period 16-JAN-09 until today, 13-FEB-09).
When I divide 2244 hours by 392 days, I get 5.72 hours per day of flying. This seems an almost impossibly high number of hours.
Anyone know anything about duty assignments at Colgan?

[wadegamm]

They (CNN) are known to cross their numbers.  It may be her total time, or her years are wrong.  As for the accident I responded this to an earlier blog:

It looks like everything happened right as they were coming up on the LOM (marker).  She was cleared, established, coming over the LOM and had just been passed off to the tower by approach.   

It looks like it is all ice related.  That is the exact point where everything would have been deploying out and the power is being slightly reduced.  The crash site is not even a half mile from the NDB-LOM ( KLUMP).   It may be the result of a partial or un-equal deployment of any one of the flaps, slats, or gear… or just a total ice function effecting lift.  The -8s only have leading edge boots and don’t have heated or bleeding wings.  I know a little Cessna can gain several hundred pounds a minute in really bad freezing rain.    As for the slight LOC (localizer) fluctuation that was reported, that is normal when another aircraft crosses the LOC runway and that is what had happened as the aircraft that made that particular report was inbound.  No one was crossing the active when 3407 was on the approach. 

I am trying to figure out which it is.  Ice effecting mechanical systems or Ice effecting the wings and weight.  There are supposed to be sensors that prevent differential deployment of flaps, slats, and speed brakes, but I don’t know the exact details of the -8 400s

 

[joeyb747]

You are correct about the Dash 8 not using bleed air in the wing anti-ice system. The only part of the ant-ice system that uses bleed air is around the engine inlets. My speculation concurs with the ice therory. Its strange that it droped out of the sky that fast and there was no mayday from the crew. If you listen to the recording, the co-pilots voice sounds relaxed in the begining, then the last transmission when they are handed off to tower, her voice sounds "hurried", but not distressed. Very strange.

[byoungblood]

My instructor and I had a chat about this on an instrument lesson today. Basically, if you're icing up, don't touch the flaps! Ice can cause such a drastic change to the wing profile that even when the flaps come down equally, it can upset the aerodynamic forces enough to upset the airplane in a hurry. In that case it is just better to carry the extra speed and land long than not at all.

When an aircraft is at or near the OM, they're only about 1600-1700 AGL. I'm not sure what approach speed they use, but as sudden as this situation developed, I doubt there was much time for recovery.

[MathFox]

There was a recent Norwegian report about an incident where an ATR 42 suddenly lost 1500 ft due to icing. http://aviation-safety.net/go.php?http://www.aibn.no/items/3216/144/8519822396/LN-FAO%20Eng..pdf

Differences with Buffalo: the Norwegian pilots had no reason to extend the flaps and they were at higher altitude.

[keith]

Being concerned about my going through my first winter with my new instrument ticket, I have been seeking out a lot of information about icing, from all sorts of sources.  I watched a great presentation on icing (I think it was by the ASF, which has lots of great material).  Two of the standout points were:
1) don't use the autopilot when experiencing icing
2) minimize configuration changes (ie. deployment of flaps).

Based on news reports in the last couple of days, it looks like both of these concepts weren't followed during this encounter.

At the very least, this can get the attention of many relatively inexperienced instrument pilots (such as myself) to hammer home these concepts.

[carrrchanged]

Someone had mentioned tailplane stall and posted this NASA video

which I added here.

I know speculation is dangerous but I think we all want to know what happened and can't help but go through possible scenarios. According to the video, almost every requirement for this dreadful event had been met including aircraft design specifics.

Only one thing was different that I could think of. The stick shaker.. They got a main wing stall. Could this have been secondary to a tail stall?

I am sorry to speculate. Also, if you were personally affected by this disaster, know that many prayers and thought are with you.

--------------
Ex regional airline CA, FE-B727   

[cimciusa]

3407 tail stall.  Posted link to NASA video focusing on tail stall and icing.
This video connects the dots. 

http://www.patricksaviation.com/videos/Contrail_25/3927/

[ogogog]

IF CGC3407 had an ice enduced tail stall then why did the aircraft pitch UP before it pitched down? if the tail stalls and looses lift which way dose the nose go? UP? the first upset dosent sound like a tail stall according to NASA.

[KSYR-pjr]

IF CGC3407 had an ice enduced tail stall then why did the aircraft pitch UP before it pitched down?

Normally the horizontal stabilizer is an airfoil that produces downward lift, meaning under typical airflow it pushes down the tail and lifts the nose.  Thus, if the horizontal stabilizer stalls it will lose its downward lift and the nose will fall.  This is why recovery for a tailplane stall is to pull UP (which reduces the angle of attack on the horizontal stab), which is opposite of the stall recovery technique for the case of the main wings stalling.

[ogogog]

that was the whole point of my question,to get people to think about what they are speculating about. in a tail stall nose goes down not up.so the question is what caused the pitch UP first?

[KSYR-pjr]

With that thought, ogogog, do you recall the Lear tail icing audio clip that originally occurred back in March 2006 and was reposted to the clips forum a few days ago? 

In that incident, the aircraft received a heavy amount of ice on its horizontal stabs (apparently that Lear model does not have any type of deicing capabilities on the stabs) while returning to Buffalo from Toronto.   That aircraft then went into an uncommanded climb for 6,000 feet before the pilots could regain control.  Fortunately the main wings didn't stall while the aircraft was climbing, as almost happened, or it would have been a pitch up followed by a violent pitch down.

I don't completely understand the physics of that particular incident.

[carrrchanged]

FYI guys & gals,
The goal of my question was not to contribute to the race of beating the NTSB to the answer. They have more facts then we are ever likely to get. And they have engineers that are smarter then (me anyway). However, the sad truth in this case is that we may never know for certain exactly what happened. The biggest problem is that we cannot catch the culprit (the ICE) red-handed. We know there was ice but how will we ever know how much and where?

But I think it is important for flight crews (and potential future ones) to go through scenarios and learn from this. Not necessarily draw conclusions but just think. Like some of you are doing here and one of the above posts suggested. I don’t know why the nose went up.. I don’t know if there was a tail stall at all.. I don’t know to what extent an autopilot could hide it if there was one.. The only thing I feel a little confident about is the stick pusher along with low IAS and high VS of impact suggests the end result was an unrecoverable main wing stall and apparently triggered by flap deployment.

They are not pleasant thoughts and they are likely to cause some anxiety next time you see ice –anywhere- on your aircraft (pilot or passenger). I know I have relatively carelessly flown through lots of it. I don’t recall any FO or CA I ever worked with explicitly turning the AP off in light or even moderate ice buildup. And what is severe? Does anyone really know? I have never reported more then light or moderate icing to ATC because to be honest I don’t know what severe looks like.

I will personally think of these Colgan crewmembers as brave, highly qualified professionals who did what they could to save a doomed airplane. This accident will likely have impact on air carrier training at some point. During my training at a number of different these, we did little more then gloss over the AE ATR accident (very similar). It was treated as an isolated case due to the ATR’s boot and flap design resulting in some AD’s. Well as of last week, this is no longer the case..

Sorry, didn't mean for this to take up so much space

[KSYR-pjr]

And what is severe? Does anyone really know? I have never reported more then light or moderate icing to ATC because to be honest I don’t know what severe looks like.

As a GA pilot who is based and flies just downwind of the US Great Lakes, I use the concept of how fast the ice accretes to determine the severity of ice (using an observation of normally a few minutes at most), since all ice accretion above a trace left to continue building would eventually become extreme.  Thus, it is how fast it builds in a few minutes, not how large the accretion eventually becomes, and believe me, I am not waiting around much past 30 seconds of observation before I am on the frequency asking for another altitude.

With that in mind, I have experienced what I consider unforecasted/unknown severe icing in both a C172 and a Bonanza V35.  In both these cases the ice accumulated to about a 1/2" inside of a couple of minutes.   My Bonanza is equipped with a TKS weeping wing (a glycol mixture leaks out of all leading edges) but ice build-up can still be seen on some unprotected parts of the aircraft. 

[flygirltammy]

Here's what Fox news is saying: "Colgan Air First Officer Rebecca Shaw, who has been with the airline since January 16, 2008, and had flown 2,244 hours with Colgan."
For the sake of discussion, let's say 16-JAN-08 is 392 days ago (365 days in a year plus 27 more for the period 16-JAN-09 until today, 13-FEB-09).
When I divide 2244 hours by 392 days, I get 5.72 hours per day of flying. This seems an almost impossibly high number of hours.
Anyone know anything about duty assignments at Colgan?

I don't know about duty day assignments at Colgan, but I used to be in pilot crew scheduling for Southwest. There are 2 things that factor in: Maximum duty as specified in the union contract (I don't know if Colgan was unionized) and maximum block time as allowed by the FAA. Maximum duty day by the union is between 8 to 12 hours (there was an additional rule at some point that the pilot had to agree to go over a certain amount of duty. I forget what the limit was), depending on what time the pilot's show time was. But the maximum SCHEDULED block time allowed by the FAA is 8 hours. And the airline can get fined by the FAA if they don't catch these legalities in time to stop the pilot from going over, like by rerouting him/her or just have them deadhead or something. The 8 hours can only be exceeded if UNSCHEDULED, which is something that would result from holds or waiting in line for departure. For Southwest, app. 6 hours of block per day would not be uncommon. I don't know if it is for Colgan.

And, I would bet money they are talking about her total time. FAA regs restrict the amount of block time for 1 year to 1,000 hours and  also 100 hours per month and 30 hours per week for commercial airline ops.

[martinez]

that`s right, the std aviation regulation of the world, icao, faa, jaa, caa. says that the top in one year, 365 days  of hour of an airline pilot is 900 hrs, 270 in 3 months and 30 in a week.

in my job in Venezuela we make 800 hrs per year in dc 9`s and in a jet is too much, in a turboprop acft is a normal ride because is a slower type of plane. but 2244 in 13 months? no way

a pilot with 2244 hrs as f/o in a dash 8? i don't think so, dash 8 is not a difficult acft .with easier systems than the dash 7 and more efficients flight controls. the normal rating to upgrade a pilot of dash 8 to the left side is almost 1.800 hrs with  recently experience in prop light acft

[carrrchanged]

I use the concept of how fast the ice accretes to determine the severity of ice (using an observation of normally a few minutes at most), since all ice accretion above a trace left to continue building would eventually become extreme.  Thus, it is how fast it builds in a few minutes, not how large the accretion eventually becomes
[/quote]

Good contribution, thanks for sharing.
Regards