SR-22 Down at Bar Harbor

Weather was right at minimums for ILS 22 - METAR KBHB 251556Z AUTO 16008KT 1 1/2SM -RA BR OVC003 17/17 A3000 RMK AO2 SLP161 P0010 T01720172
 
I was nervous when I first saw the tail number. A friend of mine has 999ET.
 
The chute cannot save you from potentially poor airmanship.... :(. There is some terrain NE which makes me....well consider carefully if I'm goign in , on that one, at mins.

RIP
 
Last edited:
Seems like a stabilized approach... however, at 5nm out on the ILS 22 (assumption) they should have been at 2100' MSL, rather than ~1000' GPS altitude.
sr22-low.jpeg
 
I spoke with a non-pilot on the scene. He stated he witnessed the Cirrus descend out of clouds, "inverted" on north end of runway. Almost the same area where the then Chief Pilot of Bar Harbor Airways, Peter Monighetti perished in a 402B along with the President & VP of Bar Harbor Airways (Caruso bros.) in similar weather conditions in May 1978. That bare last mile is tough in the fog.
 
Seems like a stabilized approach... however, at 5nm out on the ILS 22 (assumption) they should have been at 2100' MSL, rather than ~1000' GPS altitude.
View attachment 131780
Hmm, I’m reading the tabular data from Flightaware and it says the aircraft was at 2,200, then 2,000 on successive pings at N44.56 W68.29 and N44.55 W68.30, respectively. The FAF, CDLAC is at N44.55 W68.30, 6.1NM from the threshold, so I’m not seeing the same discrepancy you are.

Where are you seeing that they were at 1000’ at CDLAC? I see that he was at 1000’ much closer to the threshold, at just under 3NM.

The data from Flightradar24 and ADS-B exchange are either noisy or incomplete, so I can’t really get much there, though FR24 seems to also show they were at around 2100 at CDLAC.
 
@Flying Keys, I took the FlightAware track into FlySto.net and looked at the descent on the final approach course.

Here's a link to the interactive FlySto graphics. I got ~5 miles out by scaling up the visible runway length of 5200'.
And below a screenshot:
View attachment 131804
Ok, I don’t see a distance scale, but I measured it using Google Earth and that point is about 2.9 nautical miles from the approach end of the runway.

Figuring the GS antenna is about 900’ down the runway, he’s only about 30’ off there, well within the ADS-B margin of error.
 
@Flying Keys, I took the FlightAware track into FlySto.net and looked at the descent on the final approach course.
Are you looking at GPS altitude or Mode C pressure altitude (ref 29.92)? Both have different errors that should be accounted for before reaching any conclusions. It's labeled "MSL" which by convention implies pressure altitude, but I don't know where or how it was labeled as such. Any idea of the local altimeter setting at the time of the track?

Nauga,
who asks, "How high is up?"
 
Are you looking at GPS altitude or Mode C pressure altitude (ref 29.92)? Both have different errors that should be accounted for before reaching any conclusions. It's labeled "MSL" which by convention implies pressure altitude, but I don't know where or how it was labeled as such. Any idea of the local altimeter setting at the time of the track?

Nauga,
who asks, "How high is up?"
The data trail is as follows:
-- the aircraft broadcasts ADS-B Out which includes GPS location, ground speed, GPS altitude, etc.
-- [the aircraft in this case being a recent Cirrus writes to the SD Card a whole bunch more data... e.g. pitch, bank, IAS, altitude MSL, engine data, etc, etc. -- if this SD card survived the fire, there will be a whole bunch of answers there.]
-- ground stations run by private citizens have receivers that (if close enough nearby the aircraft) receive the ADS-B Out data and relay it to the three primary posting web sites (FlightAware, FlightRadar24, and ADSBExchange)... this is where we as observers get to review the ADS-B data... limited to GPS location, ground speed, GPS altitude, etc.

You're correct... although the tracking web sites might label it as MSL, ADS-B data is in fact GPS altitude. In this case, the display tool FlySto does label it as MSL (a bit erroneously) because FlySto also reads the full SD Card data.

Here were the METARs (altimeters ~30.0):
1) METARs:
METAR KBHB 251530Z AUTO 15005KT 1 1/2SM -RA BR OVC003 17/17 A3001 RMK AO2 P0009
METAR KBHB 251539Z AUTO 16004KT 3SM -RA BR OVC003 17/17 A3001 RMK AO2 P0009
METAR KBHB 251556Z AUTO 16008KT 1 1/2SM -RA BR OVC003 17/17 A3000 RMK AO2 SLP161 P0010 T01720172
METAR KBHB 251656Z AUTO 14004KT 1 1/2SM -RA BR OVC003 17/17 A2998 RMK AO2 SLP153 P0000 T01720172

Wayne
 
...although the tracking web sites might label it as MSL, ADS-B data is in fact GPS altitude. In this case, the display tool FlySto does label it as MSL (a bit erroneously) because FlySto also reads the full SD Card data.
ADS-B reports both pressure altitude and GPS altitude. The question is, which one is your tool using and how has it been corrected, if at all?
Where did the SD card data you refer to come from in this instance?

Nauga,
from where a miss is as good as a mile
 
Hi everyone.
Another terrible accident and we are likely that we will never find out exactly what happened.
Looking at the data there are no data points that would would point to anything way out of normal during the approach, assuming this was the intended rwy.
Looking at the plate and comparing ground references, I cannot find anything out of normal.
The last data point may indicate a slow speed for the 22 but nothing else that I can find.
I have many more data points in case anyone is interested.
Very sad.
 

Attachments

  • 990PT-CDLAC.jpg
    990PT-CDLAC.jpg
    403.8 KB · Views: 15
  • 990PT-FinalData.jpg
    990PT-FinalData.jpg
    117.9 KB · Views: 15
@nauga, I have to admit I don't know which (pressure altitude or GPS altitude) the ADS-B tracking sites (FlightAware, FlightRadar24, ADSBExchange) are using. The track display site FlySto.net in this case got a kml file output from FlightAware, so again I don't know which altitude FlightAware exported to that kml file.

The SD Card data in the panel system (Perspective+ = ~G1000) is of course in the physical panel of the crashed aircraft. I would expect it may well have been destroyed in the resulting fire. If it survived we'll likely get info from that (after 2 years) in the NTSB final report.
 
We’re focusing on correcting the altitude, but you’re still looking at the wrong place.
As @bluesideup and I pointed out, the point you selected is about halfway between the FAF (CDLAC) and the MAP, as shown by the red arrow. I backed the plane out to a point much closer to CDLAC and it’s really not off by much (and again, within a reasonable margin of error for altitude corrections).
 

Attachments

  • IMG_3638.jpeg
    IMG_3638.jpeg
    231.5 KB · Views: 16
There is a video floating around on FB of this bird coming out of the clouds stalled and spinning into the ground. Cloud deck is really low and visibility is maybe a mile. I’m guessing stalled it during the approach or maybe going missed. Last ground track speed was just 57 knots.


Sent from my iPhone using Tapatalk Pro
 
There is a video floating around on FB of this bird coming out of the clouds stalled and spinning into the ground. Cloud deck is really low and visibility is maybe a mile. I’m guessing stalled it during the approach or maybe going missed. Last ground track speed was just 57 knots.


Sent from my iPhone using Tapatalk Pro
Ugh. It’s on a YouTube short as well. I won’t link it here, but it’s not hard to find. It shows the last 1 second of the aircraft in flight, then it barely disappears behind the crown of the runway before a fireball appears.

From what I could see, the left wing was dropping and it looked like it was in an incipient spin to the left, at minimum.
 
I think we will get a detailed report of what happened with this accident. The aircraft was a g6 sr22, it had all the bells and whistles, including envelope protection, aural warnings for airspeed and glideslope , this was a well equipped aircraft. The only thing it won’t do is apply power. It also records telemetry data, along with control inputs. It does a lot.

Getting slow in a Cirrus close to the ground is a big no, no, like most aircraft.
 
  • Like
Reactions: Feh
I think we will get a detailed report of what happened with this accident. The aircraft was a g6 sr22, it had all the bells and whistles, including envelope protection, aural warnings for airspeed and glideslope , this was a well equipped aircraft. The only thing it won’t do is apply power. It also records telemetry data, along with control inputs. It does a lot.

Getting slow in a Cirrus close to the ground is a big no, no, like most aircraft.
Power management is critically important during instrument approaches with low visibility
and ceilings. It's so easy to get fixated on glideslope and localizer (or GPS centerline) and
get complacent with your airspeed. If you don't practice these low approaches regularly
(under the hood with another pilot in the right seat - preferebly at night), you're asking
for trouble in the real world. It's happens over, and over and over again.

It's tricky enough to get down to minimums with everything centered and the airspeed right
on target. But then if the runway isn't there at DH / MDA, it gets even harder on the missed
approach. The missed approach is so unnatural - it must be ingrained into our muscle memory
with the proper sequencing by repeated practice.

Lastly, these unfortunate accidents almost always aren't mysteries. Flying into low IMC
is probably one of the most unforgiving things that we can do in aviation.
 
Cirrus is very big on personal mins. Experience and more importantly currency is big in their calculations. The only way to meet their fly to actual mins personal mins allowance is to have many recent approaches in the 22 along with time and training. It’s tough to get there.

The next step up is +500 feet on da or mda and 2sm above published mins.

But these are obviously not binding.

As far as getting slow, there are many warnings the airplane gives you, like nudging the stick and a nagging “airspeed” annunciation. You even get a glide slope annunciation if you are screwing up. Unfortunately these can be turned off. The sad thing is if you are hearing them, you are the one that needs them most. I never heard them outside of training, but it was nice having them there.
 
Hi everyone.
I found out that most of the time pilots get very busy and barely live in real time, they get behind the plane.
This can only be remedied by practice, practice... and proper training.
Stay ahead, after you get properly configured for the approach your next step should be missed, if conditions are right you land.
This goes the same for VMC we should think Go around, things look good I will land... We need to make sure that pilots are ready for these transitions.
These transitions are the most difficult, you got all these things that change, Power, Attitude, Configuration, Communication..
I hope we are able to get some information and learn something from this.
 
I think we will get a detailed report of what happened with this accident. The aircraft was a g6 sr22, it had all the bells and whistles, including envelope protection, aural warnings for airspeed and glideslope , this was a well equipped aircraft. The only thing it won’t do is apply power. It also records telemetry data, along with control inputs. It does a lot.

Getting slow in a Cirrus close to the ground is a big no, no, like most aircraft.
It's also very easy to get slow on it especially on the heavier SR22 and SR22T, and it will bite you.
 
OK, my best guess. From flight track on Flightaware, it appears that he was vectored to the ILS 22. I don't have the Bangor Approach audio, can someone find the audio? It appears to be an autopilot coupled approach from the data.
At minimums he did a missed approach. (Does this airplane have a "missed approach" button on the stick?) Airplane pitched for the missed climb and power was late, airspeed deteriorated and stall spin at 200 AGL.
My guess.
 
OK, my best guess. From flight track on Flightaware, it appears that he was vectored to the ILS 22. I don't have the Bangor Approach audio, can someone find the audio? It appears to be an autopilot coupled approach from the data.
At minimums he did a missed approach. (Does this airplane have a "missed approach" button on the stick?) Airplane pitched for the missed climb and power was late, airspeed deteriorated and stall spin at 200 AGL.
My guess.
Yes, it has a TO/GA button on the throttle that activates missed approach mode on the avionics.
 
Plus, for the g6 22all you really need to do is go to full throttle after you hit the toga button if you are coupled. The airplane will get you climbing, climb rate with full flaps is 750 fpm at sea level. All you need to do after that is cleanup, set the navigation and make your call.
 
ILS 22 has a note “Autopilot coupled approach NA below 900 feet MSL”, wondering what’s that about. Unreliable GS?
 
ILS 22 has a note “Autopilot coupled approach NA below 900 feet MSL”, wondering what’s that about. Unreliable GS?

Would the crossing altitude restriction at JESUL (at or above 780’ have anything to do with it? I see an 303ft obstacle right in the glide path. All minimums are well below the 900MSL note.
 
Prelim NTSB report came out, wreckage was found left of the approach end of Rwy 22, witness in the field reported thick fog conditions at the time and saw the plane approaching, going full power and observed it descending rapidly on a left handed turn.

 
Back
Top