Pilatus PC-12 down near San Angelo,TX

Josey Wales

Filing Flight Plan
Joined
Aug 26, 2020
Messages
29
Display Name

Display name:
Josey Wales
A friend sent this to me last night. It appears to have gone down while flying an instrument approach into KSJT. Conditions in the area at the time were overcast with ceilings around 1200 AGL. Not sure of exact tips at the time, but when I checked earlier in the day, the tops and freezing level were both around 10k MSL.

My condolences to the families.

 
Just listened to the audio, they reported over shooting the final for 36, and that they had a problem.
 
Dang, I’m sorry to hear that.

There will be a lot to unpack with this one.

For a timestamp, the last ADS-B hit was on 12/14 at 1817:47Z

Here was the last METAR at SJT:

KSJT 141751Z 10006KT 10SM OVC011 10/08 A3031 RMK AO2 SLP263 60004 T01000083 10100 20072 58013

No significant changes at the next routine observation, just a slightly better ceiling

The TAF for that time period was:

KSJT 141737Z 1418/1518 11010KT P6SM OVC020

The next change wasn’t forecast until 0200. SJT was issuing new TAFs every hour or two for a while, indicating lower ceilings.
Winds aloft at 3000’ were 150 at 21
Winds aloft at 6000’ were 179 at 33

Airport elevation is 1919

N188PC was on a vector to DBASE, which is the center IAF at the T-style RNAV platform (has a HILPT). The profile altitude is 4100’ minimum until crossing DBASE, then 4000’ to TIMPP.

According to ADS-B data, 8PC was approaching DBASE from the northeast on a 230° heading, with autopilot and alt hold on at 4300 selected, ground speed 190.

(communication times approximate and comms paraphrased, ADS-B times from the website)

(Edit: at some point 8PC was given a 210 heading, but this is not apparent in comms)

At 1816:30, 8PC was 3NM SE of DBASE on a 210° heading, still 4300, autopilot and alt hold. ATC gave them a left turn to 270, 8PC asked if that was supposed to be right, quipped that left is the long way around” ATC confirmed right 270

In the turn to 270, groundspeed increased to 230kts, no other changes

At 1817:14, ATC gave the PTAC: 8 miles from TIMPP, turn right 030(!) maintain 4300 until established on the final approach course, cleared RNAV 36.

I want to interject here that to this point, in the last 17 minutes of the recording twice I heard ATC tell 8PC 4000 instead of 4300, and the first time, several minutes prior, 8PC also read back 4000 (thus heard what I heard). In playing it back, the controller veritably mumbled “three hundred” both times, kind of fast talk and tailed off. It was hard to discern. Anyway, 8PC read back the approach clearance correctly.

Per ADS-B, at 1817:30, the heading selector was set to 300°(!) altitude 4300 in a slight descent and groundspeed 241

Just about the same time, 8PC transmitted “we overshot it” ATC responded, “that’s why I gave you 030.”

At about 1817:38, the autopilot mode was changed to LNAV and alt hold. The aircraft was on a 320° track and about 450fpm descent, which increased to 1000fpm a few seconds later. FMS SEL still showed about 4300’ selected on the alt hold as they descended through about 3800’

5 seconds later, the vertical speed went over 4000, increasing to 5000 down. No apparent change from the FMS.

About that same time, 8PC said “we got a problem”
That was the last transmission from 8PC

The last ADS-B hit was four seconds later,
At 1817:47. 259kts GS, 2800’ -5056 fpm.
No apparent change in the autopilot. The ground in that area is about 2100’ MSL.

I’m still chewing on this, but those are the approximate data so far.

Sad, RIP all.

Edit to remove reference to the airmet and CWA which were a couple dozen miles to the NW.
 
Last edited:
One thing that strikes me is how the groundspeed went from 190 to 230 after a 40° heading change in what should have been level flight.

Edit: after doing a bunch of math, it gives me a TAS of around 219 kts and a wind of 163@47, which was higher than forecast. Will continue to check if there are more solutions that work. (refining a bit more)

Last edit for this one: I figured out why my first calculation was off. The LiveATC playback missed a turn to 210° around 1814z (I think the scanner was tuned to another channel). That makes it a 60° turn to 270, not 40° as I originally thought. So the groundspeed shift solution is not as dramatic - easily explainable by strong, but not extreme winds aloft.
 
Last edited:
Dang, I’m sorry to hear that.

There will be a lot to unpack with this one.

For a timestamp, the last ADS-B hit was on 12/14 at 1817:47Z

Here was the last METAR at SJT:

KSJT 141751Z 10006KT 10SM OVC011 10/08 A3031 RMK AO2 SLP263 60004 T01000083 10100 20072 58013

No significant changes at the next routine observation, just a slightly better ceiling

The TAF for that time period was:

KSJT 141737Z 1418/1518 11010KT P6SM OVC020

The next change wasn’t forecast until 0200. SJT was issuing new TAFs every hour or two for a while, indicating lower ceilings.
Winds aloft at 3000’ were 150 at 21
Winds aloft at 6000’ were 179 at 33

Airport elevation is 1919

N188PC was on a vector to DBASE, which is the center IAF at the T-style RNAV platform (has a HILPT). The profile altitude is 4100’ minimum until crossing DBASE, then 4000’ to TIMPP.

According to ADS-B data, 8PC was approaching DBASE from the northeast on a 210° heading, with autopilot and alt hold on at 4300 selected, ground speed 190.

(communication times approximate and comms paraphrased, ADS-B times from the website)
At 1816:30, 8PC was 3NM SE of DBASE, still 120°, still 4300, autopilot and alt hold. ATC gave them a left turn to 270, 8PC asked if that was supposed to be right, quipped that left is the long way around” ATC confirmed right 270

In the turn to 270, groundspeed increased to 230kts, no other changes

At 1817:14, ATC gave the PTAC: 8 miles from TIMPP, turn right 030(!) maintain 4300 until established on the final approach course, cleared RNAV 36.

I want to interject here that to this point, in the last 17 minutes of the recording twice I heard ATC tell 8PC 4000 instead of 4300, and the first time, several minutes prior, 8PC also read back 4000 (thus heard what I heard). In playing it back, the controller veritably mumbled “three hundred” both times, kind of fast talk and tailed off. It was hard to discern. Anyway, 8PC read back the approach clearance correctly.

Per ADS-B, at 1817:30, the heading selector was set to 300°(!) altitude 4300 in a slight descent and groundspeed 241

Just about the same time, 8PC transmitted “we overshot it” ATC responded, “that’s why I gave you 030.”

At about 1817:38, the autopilot mode was changed to LNAV and alt hold. The aircraft was on a 320° track and about 450fpm descent, which increased to 1000fpm a few seconds later. FMS SEL still showed about 4300’ selected on the alt hold as they descended through about 3800’

5 seconds later, the vertical speed went over 4000, increasing to 5000 down. No apparent change from the FMS.

About that same time, 8PC said “we got a problem”
That was the last transmission from 8PC

The last ADS-B hit was four seconds later,
At 1817:47. 259kts GS, 2800’ -5056 fpm.
No apparent change in the autopilot. The ground in that area is about 2100’ MSL.

I’m still chewing on this, but those are the approximate data so far.

Sad, RIP all.

Edit to remove reference to the airmet and CWA which were a couple dozen miles to the NW.
How can ADS-B know how the autopilot was set? Or what the heading selector was set on?
 
How can ADS-B know how the autopilot was set? Or what the heading selector was set on?
Some new avionics systems can retrieve navigation modes from the FMS and broadcast it in their ADS-B out.
 
Where does it go? How long is the information held wherever that is?
If it’s being tracked by ADS-B exchange (as this was), then however long they retain their data. How long it’s available to the public to view is another story, but I can see what it reported as the a/p mode selection, altitude, and heading, as clear as day.
 
Controller was behind, the first turn to final was a right turn to 030 for a 002 final approach course.

Read my summary. The aircraft was doing 190 knots over the ground, the controller turned him 40° 60° right to a 270 heading prior to initiating the PTAC and the ground immediately speed went up 40 knots, plus another 20 less than a minute later. The PTAC included the 030 heading and the pilot read it back. However, the heading hold was set to 300.

The pilot said “we overshot it”
The controller responded “I know, that’s why I gave you a 030.”

Could the controller have clarified he was bringing him through and back on the other side? I’ll let the controllers weigh in. But my gut is reading some expectation bias into it, which caused confusion. The solution to shoot through and back from the west side of final was exacerbated by the high groundspeed.
 
Last edited:
I studied all this audio and the ADS-B data very late last night. I'm not a pilot, just enjoy learning about aviation. I live a few miles from where this happened and I run an ADSB feeder to flightaware (which incidentally masked this flight) and a broadcastify.com air band feeder. ( In contrast with live ATC, my broadcastify channel has both FW Center as well as SJT approach and tower, so, the live ATC dotcom audio is more complete than broadcastify, as center is always pretty busy and my scanner can't hear both at once.)

I noticed a couple of things studying the audio and ADS-B, that are probably nothing. But I hope to learn from the answers and responses to my comments:

1) the ADS-B data from flightaware contains a pressure reading. Not being a pilot, I am curious if that reading can be worked backwards to determine if his altimeter was or was not set correctly?

2) I'm not saying this was or wasn't a factor, but just giving some local eyeball data: there is a wind turbine farm and enormous transmission lines serving that windfarm, just to the south and west of the last ADSB transmissions.

Thanks.
 
Last edited:
Read my summary. The aircraft was doing 190 knots over the ground, the controller turned him 40° 60° right to a 270 heading prior to initiating the PTAC and the ground immediately speed went up 40 knots, plus another 20 less than a minute later. The PTAC included the 030 heading and the pilot read it back. However, the heading hold was set to 300.

The pilot said “we overshot it”
The controller responded “I know, that’s why I gave you a 030.”

Could the controller have clarified he was bringing him through and back on the other side? I’ll let the controllers weigh in. But my gut is reading some expectation bias into it, which caused confusion. The solution to shoot through and back from the west side of final was exacerbated by the high groundspeed.
He did read back 030. But it looks like he set the bug at 300. It’s about 10 seconds later, like about 3 degrees per second, he says he has a problem. I think a good guess is he was startled by the plane stopping turn at 300. Question would be, how did he react
 
Controller was behind, the first turn to final was a right turn to 030 for a 002 final approach course.
The more I analyze this, the more I tend to agree with your assessment.
He did read back 030. But it looks like he set the bug at 300. It’s about 10 seconds later, like about 3 degrees per second, he says he has a problem. I think a good guess is he was startled by the plane stopping turn at 300. Question would be, how did he react
Also agree.
 
I studied all this audio and the ADS-B data very late last night. I'm not a pilot, just enjoy learning about aviation. I live a few miles from where this happened and I run an ADSB feeder to flightaware (which incidentally masked this flight) and a broadcastify.com air band feeder. ( In contrast with live ATC, my broadcastify channel has both FW Center as well as SJT approach and tower, so, the live ATC dotcom audio is more complete than broadcastify, as center is always pretty busy and my scanner can't hear both at once.)

I noticed a couple of things studying the audio and ADS-B, that are probably nothing. But I hope to learn from the answers and responses to my comments:

1) the ADS-B data from flightaware contains a pressure reading. Not being a pilot, I am curious if that reading can be worked backwards to determine if his altimeter was or was not set correctly?

2) I'm not saying this was or wasn't a factor, but just giving some local eyeball data: there is a wind turbine farm and enormous transmission lines serving that windfarm, just to the south and west of the last ADSB transmissions.

Thanks.

1. The data show his altimeter was set to 30.30, which is close to the observation at the time. Regardless of the data, if it was set incorrectly, the controller had a lot of time to let him know if he was off. Also, there weren't any low altitude alerts issued.

2. The ground is at about 2100 and the windfarm tops out at 3000' (and was a few miles the south of his turn). The loss of control seems to have started well before any terrain or obstacles became an issue.

It's always good to check into altimeter issues, but I don't believe they were a factor here.
 
1. The data show his altimeter was set to 30.30, which is close to the observation at the time. Regardless of the data, if it was set incorrectly, the controller had a lot of time to let him know if he was off. Also, there weren't any low altitude alerts issued.

2. The ground is at about 2100 and the windfarm tops out at 3000' (and was a few miles the south of his turn). The loss of control seems to have started well before any terrain or obstacles became an issue.

It's always good to check into altimeter issues, but I don't believe they were a factor here.
Thanks. The ADS-B data had the pressure data in a different unit of measure, and it wasn't obvious to me how to do the conversion. When I did the conversion using an online calculator it seemed to be a couple inches mercury off.

The ADS-B data read "QNH 1026.4 hPa" ; I tried making that calculation and kept coming up with like 32 inches.
 
Legacy PC12 autopilots don’t handle turbulence well, and tend to kick off in anything more than light. I am wondering if this is another case of a capable airplane flown poorly by someone without to skills to match the automation.

The PC12 is an incredibly capable airplane, and very easy to fly. If poorly managed it will also get away from you in a big hurry.

This accident is a good example of what I think is a similar situation.
 
Last edited:
He did read back 030. But it looks like he set the bug at 300. It’s about 10 seconds later, like about 3 degrees per second, he says he has a problem.
Judging from the inflection in voice and the increased background noise, I think control had already been lost when that transmission was made.
 
If it’s being tracked by ADS-B exchange (as this was), then however long they retain their data. How long it’s available to the public to view is another story, but I can see what it reported as the a/p mode selection, altitude, and heading, as clear as day.
As the founder of ADDBexchange, I can confirm - we don’t ever delete the data.
 
Legacy PC12 autopilots don’t handle turbulence well, and tend to kick off in anything more than light. I am wondering if this is another case of a capable airplane flown poorly by someone without to skills to match the automation.

The PC12 is an incredibly capable airplane, and very easy to fly. If poorly managed it will also get away from you in a big hurry.

This accident is a good example of what I think is a similar situation.
Link was TLDR, tonight, maybe another time. You said “tend to kick off…” If it suddenly kicked off, could it have been way out of trim when that happened?1
As the founder of ADDBexchange, I can confirm - we don’t ever delete the data.
can you post a screen shot or link to this incident? I don’t know a lot about it but I’ve seen a lot of shots of it in threads like this. I cannot recall seeing anything about what the avionics in the plane were set at. What I see is a time, an altitude, speed I think, direction. By direction I mean there is a number corresponding to a 360 degree circle. I always assumed it to be Track. That it could know what the heading was, or what a pilot had set on the bug didn’t pass the logic test with me.
 
Link was TLDR, tonight, maybe another time. You said “tend to kick off…” If it suddenly kicked off, could it have been way out of trim when that happened?

No. If the autopilot kicks off the airplane will be (more or less) trimmed for the airspeed it is currently at.
 
Link was TLDR, tonight, maybe another time. You said “tend to kick off…” If it suddenly kicked off, could it have been way out of trim when that happened?1

can you post a screen shot or link to this incident? I don’t know a lot about it but I’ve seen a lot of shots of it in threads like this. I cannot recall seeing anything about what the avionics in the plane were set at. What I see is a time, an altitude, speed I think, direction. By direction I mean there is a number corresponding to a 360 degree circle. I always assumed it to be Track. That it could know what the heading was, or what a pilot had set on the bug didn’t pass the logic test with me.

Look at the left hand side of the screen and scroll down the left hand toolbar.
 
Link was TLDR, tonight, maybe another time. You said “tend to kick off…” If it suddenly kicked off, could it have been way out of trim when that happened?1

can you post a screen shot or link to this incident? I don’t know a lot about it but I’ve seen a lot of shots of it in threads like this. I cannot recall seeing anything about what the avionics in the plane were set at. What I see is a time, an altitude, speed I think, direction. By direction I mean there is a number corresponding to a 360 degree circle. I always assumed it to be Track. That it could know what the heading was, or what a pilot had set on the bug didn’t pass the logic test with me.

1702710483120.png
 
FYI, when you bring up the flight, you can press the “K” button and see each of the track points. If you click on them, data from that track point will be shown.
 
This is what I get when click on that link..

View attachment 123319
Looks like you are using an Apple iPad, I can't really assist on that as far as how to see what I see. On chrome browser, the lower left hand box contains an up-down scroll bar and I can view all parameters received.
 
Looks like you are using an Apple iPad, I can't really assist on that as far as how to see what I see. On chrome browser, the lower left hand box contains an up-down scroll bar and I can view all parameters received.
Thx. So the browser could be the difference. I’ll try it on a laptop where I have Firefox or Edge to try when I get to it.
 
Thx. So the browser could be the difference. I’ll try it on a laptop where I have Firefox or Edge to try when I get to it.
That window on the bottom left can scroll and show the same info. At lease it does on my iPhone
 
Back
Top