Airlines PNG Dash 8 fatal crash in PNG
Moderator
I've never flown a Dash 8, but I have fair experience in DHC-6 and Caravan. On both those types, rapidly moving the propeller levers from coarse pitch to fine pitch can get you close to an overspeed, though 60% over seems very hard to imagine. The key is that power lever movement alone should not really result in significant overspeed, as the governor should govern. I speculate that overspeed risk is cause by mishandling the propeller levers.
What would happen if the propeller controls were moved toward feather in error, and then rapidly moved back to fine to correct things, while lots of power was applied? Could the governor change the blade angle quickly enough to prevent an overspeed?
Though I respect what Bombardier would instruct about the operation of the Dash 8, I would speculate that if a pilot selected power below flight idle in flight, that would initially result in a lower propeller speed, as there would not be much engine power commanded to turn the discing propellers. The discing propellers would have some of the blade in reverse pitch, thus large torque required to turn the prop.
Just me thinking, I have no knowledge of the accident....
What would happen if the propeller controls were moved toward feather in error, and then rapidly moved back to fine to correct things, while lots of power was applied? Could the governor change the blade angle quickly enough to prevent an overspeed?
Though I respect what Bombardier would instruct about the operation of the Dash 8, I would speculate that if a pilot selected power below flight idle in flight, that would initially result in a lower propeller speed, as there would not be much engine power commanded to turn the discing propellers. The discing propellers would have some of the blade in reverse pitch, thus large torque required to turn the prop.
Just me thinking, I have no knowledge of the accident....
Join Date: Oct 2006
Location: Antigua, W.I.
Posts: 127
Likes: 0
Received 0 Likes
on
0 Posts
In discing the blade angle decreases from 17.5 to a minimum of 1.5 degrees. However, there is no significant change in Tq.
The Bombardier AOM states:
"When the POWER Levers are moved aft of the FLIGHT IDLE gate in-flight, the Propeller Governor, Propeller Overspeed Governor and the BETA Backup logic protection are all inhibited
and Propeller speed control is no longer available. In this condition the propeller(s) would be
driven uncontrollably toward a Reverse pitch condition resulting in an overspeeding propeller
and substantial engine damage leading to possible engine failure."
However, if this was the case, I suspect the plane would come straight down. I've never heard of a prop overspeed of 60% due to Condition Lever(s) being moved too rapidly and I suspect if such a thing did occur, the manufacturer would send out a crew advisory.
The Bombardier AOM states:
"When the POWER Levers are moved aft of the FLIGHT IDLE gate in-flight, the Propeller Governor, Propeller Overspeed Governor and the BETA Backup logic protection are all inhibited
and Propeller speed control is no longer available. In this condition the propeller(s) would be
driven uncontrollably toward a Reverse pitch condition resulting in an overspeeding propeller
and substantial engine damage leading to possible engine failure."
However, if this was the case, I suspect the plane would come straight down. I've never heard of a prop overspeed of 60% due to Condition Lever(s) being moved too rapidly and I suspect if such a thing did occur, the manufacturer would send out a crew advisory.
It seems that at least one prop (From one of the the preliminary report Photos) is feathered, so how can you get 60% over speed on a feathered prop ? I'm not a -8 Driver but it just puzzles me... 28 deaths, I really want to find out what happened !
Join Date: Nov 2009
Location: USA
Posts: 74
Likes: 0
Received 0 Likes
on
0 Posts
Here is a clue:
http://www.airdisaster.com/reports/ntsb/AAR94-06.pdf
Not saying this is what happened here, but way back in '94 a Saab 340, for some reason (read the report) ended up with both engines in reverse range in flight. Both props ran away; one going to 142%, the other to 158%. Both engines flamed out, but they were able to get the engines running again.
Didn't help, though. In a free turbine engine there is no direct mechanical connection between the compressor (gas generator) section and the power section (which drives the propeller). While the overspeed event did not damage the compressor sections, it destroyed the power turbines in both engines. So even though both were successfully restarted, with the power turbine blown out the back like a mouthful of sunflower seed husks, there was no way to deliver power to the propellers.
Unlike the accident in PNG, there were no serious injuries among the pax and crew of the Saab. And again, I am not saying this is what happened. but there are similarities between the preliminary report on the PNG crash and the NTSB report referenced above.
http://www.airdisaster.com/reports/ntsb/AAR94-06.pdf
Not saying this is what happened here, but way back in '94 a Saab 340, for some reason (read the report) ended up with both engines in reverse range in flight. Both props ran away; one going to 142%, the other to 158%. Both engines flamed out, but they were able to get the engines running again.
Didn't help, though. In a free turbine engine there is no direct mechanical connection between the compressor (gas generator) section and the power section (which drives the propeller). While the overspeed event did not damage the compressor sections, it destroyed the power turbines in both engines. So even though both were successfully restarted, with the power turbine blown out the back like a mouthful of sunflower seed husks, there was no way to deliver power to the propellers.
Unlike the accident in PNG, there were no serious injuries among the pax and crew of the Saab. And again, I am not saying this is what happened. but there are similarities between the preliminary report on the PNG crash and the NTSB report referenced above.
There is no feathering springs in the prop hubs. Piston inside with engine oil on both sides, the system is controlled by a hydraulic (engine oil) propeller control unit (PCU), oil is ported in and out of the prop via a transfer tube.
Originally Posted by BobnSpike
Here is a clue:
Originally Posted by twochai
The HS prop on any PW 100 series engine will be driven to feather after loss of oil pressure by springs after any normal, or abnormal shutdown.
Originally Posted by Livesinafield
Possibly just feathered the prop after the overspeed causing a loss of control
Bobnspike, The AE accident relates to a different engine type and systems. I'm not sure this is what happened here. In that case the Capt may have selected reverse beta, just like the Twin Otter crash in St Barts.
French BEA report (french) here: http://aviation-safety.net/go.php?ht...f-es010324.pdf
BEA english report: http://www.bea.aero/docspa/2001/f-es...-es010324a.pdf
ASN page : ASN Aircraft accident de Havilland Canada DHC-6 Twin Otter 300 F-OGES Saint-Barthlmy
French BEA report (french) here: http://aviation-safety.net/go.php?ht...f-es010324.pdf
BEA english report: http://www.bea.aero/docspa/2001/f-es...-es010324a.pdf
ASN page : ASN Aircraft accident de Havilland Canada DHC-6 Twin Otter 300 F-OGES Saint-Barthlmy
Last edited by Melax; 13th Nov 2011 at 13:41.
Join Date: Nov 2004
Location: Here, there, and everywhere
Posts: 1,122
Likes: 0
Received 12 Likes
on
7 Posts
If not approved, placing power levers below flight idle on a turboprop can be very dangerous. The F50 has a different propeller(Dowty) for its PW100 engine but as this accident report shows, the three second time period that it was done on this aircraft was fatal. Do a 360(or some other safe manouver) if you are too high.
http://www.skybrary.aero/bookshelf/books/677.pdf
http://www.skybrary.aero/bookshelf/books/677.pdf
Join Date: Dec 2000
Location: On the equator
Posts: 1,291
Likes: 0
Received 0 Likes
on
0 Posts
From the preliminary report, they were at 13,000 ft with 24 miles to run before Madang. This seems to be quite high. At what altitude should you be at 24 miles for a normal descent profile (assuming already clear of terrain)?
Join Date: May 2004
Location: Bear Island
Posts: 598
Likes: 0
Received 0 Likes
on
0 Posts
FL80 or thereabouts ... depending on the platform altitude /entry point for the procedure ... and of course terrain in that sector. The above assumes terrain is not an issue.
Join Date: Nov 2009
Location: USA
Posts: 74
Likes: 0
Received 0 Likes
on
0 Posts
The CT-7's and PW-100's both have a free power turbine. Additionally, in both engines, coming over the gates into beta or reverse disables the propeller overspeed protection. The preliminary report states that according to the FDR on the Dash, the props went to 160%, which implies the overspeed protection was not operating.
Even with the Dash's good descent performance, 13000 at 24 miles is way high. I am not familiar with the terrain in PNG, but discounting terrain issues, a standard descent ratio is about 3:1, making the correct altitude at 24 NM 7200 ft.
Even with the Dash's good descent performance, 13000 at 24 miles is way high. I am not familiar with the terrain in PNG, but discounting terrain issues, a standard descent ratio is about 3:1, making the correct altitude at 24 NM 7200 ft.
Join Date: Nov 2011
Location: Stranded
Posts: 67
Likes: 0
Received 0 Likes
on
0 Posts
If the propellers were placed into DISC (retarded below the flight idle indent) it is possible the aerodynamic forces of a rapid descent could cause the propellers to overspeed. In normal flight configuration the DHC-8 has an automatic safety feature that will drive the propeller to coarse pitch if it exceeds 1208 PRPM. This is driven by engine oil pressure pushing on either side of a piston and tachometer. In the oil is lost the propeller will lock its pitch.
Bombardier already has issued warnings about reducing the power levers below flight idle and in US air carriers a placard is mandatory stating this prohibition.
As a manager at another DHC-8 operator in the Pacific I will be interested in seeing how this investigation plays out.
Bombardier already has issued warnings about reducing the power levers below flight idle and in US air carriers a placard is mandatory stating this prohibition.
As a manager at another DHC-8 operator in the Pacific I will be interested in seeing how this investigation plays out.
Join Date: Oct 2006
Location: Antigua, W.I.
Posts: 127
Likes: 0
Received 0 Likes
on
0 Posts
Actually Island-Flyer, the Bombardier AOM says that in flight selection below Flight Idle the "props would be driven uncontrollably to a reverse pitch condition"
This would probably overspeed the power turbine resulting in failure. If the crew in such a situation was lucky enough to get the props out of reverse, the only option left - with the power turbine failed - would be to feather the prop(s)
This would probably overspeed the power turbine resulting in failure. If the crew in such a situation was lucky enough to get the props out of reverse, the only option left - with the power turbine failed - would be to feather the prop(s)
Join Date: Aug 2009
Location: earth
Posts: 33
Likes: 0
Received 0 Likes
on
0 Posts
A standard 3x profile on a Lae-Madang track will not work, bcoz of the high terrain you have to clear and Madang is just across the bay after you've cross those big rocks.
Just a quick question, does the Dash8 do overspeed test on there run ups like they do in the Otter? I've only seen Dash do Auto FX test on there run ups?
Just a quick question, does the Dash8 do overspeed test on there run ups like they do in the Otter? I've only seen Dash do Auto FX test on there run ups?
Join Date: Jan 2008
Location: Nairobi
Age: 46
Posts: 23
Likes: 0
Received 0 Likes
on
0 Posts
In response to the question on the overspeed governor on the DHC-8, the overspeed governor test is now an engineering fuction and is not carried out by the flight crews on the normal first flight of the day checks.
Slight thread drift here, but similarly the T-handles are not pulled to test the fuel shut-off valves, we only test the loops for continuity, and the fire bells and fire detect lights in our company. This has also been relegated to an engineering fuction.
Slight thread drift here, but similarly the T-handles are not pulled to test the fuel shut-off valves, we only test the loops for continuity, and the fire bells and fire detect lights in our company. This has also been relegated to an engineering fuction.
Join Date: Nov 2009
Location: USA
Posts: 74
Likes: 0
Received 0 Likes
on
0 Posts
I figured terrain would be an issue.
13000 feet AGL at 24 NM is 540 ft/NM. That's pretty steep. At 180 knots groundspeed that would require just over 1600 ft/min descent rate. At 240 it would require just under 2200. Distance required to slow and configure would further steepen the required descent angle.
Perhaps someone with experience in the Dash can let us know whether those descent rates are practical.
13000 feet AGL at 24 NM is 540 ft/NM. That's pretty steep. At 180 knots groundspeed that would require just over 1600 ft/min descent rate. At 240 it would require just under 2200. Distance required to slow and configure would further steepen the required descent angle.
Perhaps someone with experience in the Dash can let us know whether those descent rates are practical.