Plane breaks up mid air over Long Island

The autopilot is your friend, provided it's all electric.
 
Ya know, having stratus w/ foreflight could be a game changer in this situation too. And for very low cost.
It has an AI.
 
Anyone doing their instrument training should spend some time in the sim with a good instructor.

One of the early things my II had be do was "just fly the attitude indicator" as he failed the vac system, puts you into a text book grave yard spiral.


Mentally being able decide a instrument and or system has gone FUBAR and then completely ignore it is not the easiest thing to do.

RIP
 
ifly4fun said:
Theres also the big assumption that the pilot recognizes they are disorientated to begin with. For all we know he thought he was straight and level and recovering nicely.
Click to expand...
No, if you've ever really been there you feel the G and know something isn't right. And just like quicksand the harder you fight the quicker you die.
 
ifly4fun said:
Theres also the big assumption that the pilot recognizes they are disorientated to begin with. For all we know he thought he was straight and level and recovering nicely.
Click to expand...

Yeah, recognition is the scary part in stories like this. It's entirely possible that his first recognition of being in an unusual attitude was the breaking of the airplane.
 
Art VanDelay said:
No, if you've ever really been there you feel the G and know something isn't right. And just like quicksand the harder you fight the quicker you die.
Click to expand...

Indeed, not unloading the plane often results in a snap, reminds me of the private piloted PC12, autopilot came off, dude spent much of his focus on the AP, once he realized he was in a spiral he was well past VNE, he yanked, the PC12 snapped.

pc-12-crash-site.jpg
 
James331 said:
Anyone doing their instrument training should spend some time in the sim with a good instructor.

One of the early things my II had be do was "just fly the attitude indicator" as he failed the vac system, puts you into a text book grave yard spiral.


Mentally being able decide a instrument and or system has gone FUBAR and then completely ignore it is not the easiest thing to do.

RIP
Click to expand...
Real partial panel is way harder than covering up instruments with post-its or those black things. Those first few moments when the attitude indicator starts showing a climbing turn (usually what happens) are very disorienting until you realize cross-checks are not verifying and declare the institment FUBAR. Then if you don't physically cover up the broken insturments it's VERY distracting seeing them in the scan even if you try not to look at it.

This is all made clear in a good sim, but they should almost have IFR trainers where an instructor can actually fail the vac system and see how the pilot reacts... not just put a post-it on it and say 'whoops looks like that failed.'

What's interesting in this situation relative to the normal training scenario is the vac failure (what the pilot reported so TBD if that's what happened) occurred in VMC... then the aircraft entered IMC when the pilot already know things were not working right.
 
James331 said:
Anyone doing their instrument training should spend some time in the sim with a good instructor.

One of the early things my II had be do was "just fly the attitude indicator" as he failed the vac system, puts you into a text book grave yard spiral.


Mentally being able decide a instrument and or system has gone FUBAR and then completely ignore it is not the easiest thing to do.

RIP
Click to expand...

I recently flew the Redbird FMX with the Bonanza panel. The instructor gave an unannounced, un-briefed vacuum failure. I was amazed at how long it took me to recognize I was in trouble. And yes, it does take a little practice to recognize and recover from the death spiral.

One of the best analogies I've heard was to liken the graveyard spiral to the little coin "funnels" you see, where the coin circles faster and faster as it spirals down into the narrower end. That's exactly what happens when you pull back on the yoke without first leveling the wings.
 
rocketflyer84 said:
Autopilot is really your friend in this situation.
Click to expand...
If it works. With a failing panel, it might not work. The more sophisticated autopilots have more ways to fail. For instance, a GFC700 will give it up if your airspeed or attitude is out of bounds, because it thinks the input is bad. When it isn't putting you in an unusual attitude on its own (yes, I had one do that to me, fortunately in VMC at 4000 AGL).

Simpler wing levelers like some of the one axis STECs depend only on a turn coordinator, but you'll have to manage the dive on your own. To correct pitch, you'll need either an AI interface (dead on the accident airplane) or an altimeter interface (can get fooled bad by static port ice).
 
rocketflyer84 said:
Autopilot is really your friend in this situation.
Click to expand...

Provided you understand your autopilot. I would bet that many pilots with a Cessna 300A don't understand that in NAV and HDG mode it is vacuum-dependent, but in wing leveler mode it is only electric-dependent (driven from the TC).
 
ChemGuy said:
Remember covers or post it's to hide those dead instruments. Distracting as hell to see them present wrong info right in the middle of your scan.
Click to expand...

Yeah, this is so true. Worst I've ever had is a tumbling AI and it was shockingly difficult to ignore until I found a piece of scrap paper and hung it over the bad instrument.
 
ChemGuy said:
Sad to hear about this. Aluminum showers are never good.
Sounds almost like some get-there-itis at work. Instead of biting the bullet and flying back/toward VMC they stayed where there was heavy OVC with failing panel.
Click to expand...
You are assuming that he departed VFR, and wasn't IFR rated. I don't think that is established. He may have departed IFR, climbed through IMC, and then lost his panel.
 
172andyou said:
Yeah, recognition is the scary part in stories like this. It's entirely possible that his first recognition of being in an unusual attitude was the breaking of the airplane.
Click to expand...

This is why I asked my question above about when to pull the chute. By the time you recognize you're in trouble, is it too late?
 
PPC1052 said:
You are assuming that he departed VFR, and wasn't IFR rated. I don't think that is established. He may have departed IFR, climbed through IMC, and then lost his panel.
Click to expand...
I don't see that.

Even for a rated and current instrument pilot, electing to enter IMC with a partial panel when alternatives exist is questionable judgement. Yes, we're trained to fly with a dead vacuum pump, but it is always safer to remain in VMC if the option exists.
 
MAKG1 said:
If it works. With a failing panel, it might not work. The more sophisticated autopilots have more ways to fail. For instance, a GFC700 will give it up if your airspeed or attitude is out of bounds, because it thinks the input is bad. When it isn't putting you in an unusual attitude on its own (yes, I had one do that to me, fortunately in VMC at 4000 AGL).

Simpler wing levelers like some of the one axis STECs depend only on a turn coordinator, but you'll have to manage the dive on your own. To correct pitch, you'll need either an AI interface (dead on the accident airplane) or an altimeter interface (can get fooled bad by static port ice).
Click to expand...

What's wrong with airspeed?
 
SixPapaCharlie said:
Ya know, having stratus w/ foreflight could be a game changer in this situation too. And for very low cost.
It has an AI.
Click to expand...

I was out flying yesterday as a passenger and we were in and out of imc. I had my stratus suction cupped to the side window and calibrated it on the ground before take off. On taxi it started to show a dive. I recalibrate it and while flying that thing was again showing a dive and a turn while straight and level. Stratus was fully charged too..not sure I would trust the stratus AHRS. I want to and that's why I went with the 2 for backup but it doesn't seems to be reliable. And this isnt the first time it has done it. Actually ive never had the ahrs work right...Ugh..
 
Last edited:
MAKG1 said:
I don't see that.

Even for a rated and current instrument pilot, electing to enter IMC with a partial panel when alternatives exist is questionable judgement. Yes, we're trained to fly with a dead vacuum pump, but it is always safer to remain in VMC if the option exists.
Click to expand...

Very widespread cloud deck yesterday. Even with 2+ hours of fuel, he may not have been able to get down to vfr without dropping through the deck at some point.
No one's mentioned the possibility of an iced up pitot tube. Freeze level up there today is 4k, not much different conditions than yesterday, or earlier in the week for that matter. In the confusion, forgetting to turn on pitot heat...recipe for disaster.
 
MAKG1 said:
I don't see that.

Even for a rated and current instrument pilot, electing to enter IMC with a partial panel when alternatives exist is questionable judgement. Yes, we're trained to fly with a dead vacuum pump, but it is always safer to remain in VMC if the option exists.
Click to expand...
thats what i was getting at. Even if rated and current, going into IMC for more than a quick thru the deck, with a know panel issue, and possibly more failures or pilot issues was a bad idea. But if that VMC weather was an hour in the wrong direction and he couldnt spend the time or money to go that way, then i would definitely say get-there-itis played a role.
Unless there was no reachable VMC or scattered/broken layers to go thru. Then you just have to tighten up, fly right and punch thru.
 
MAKG1 said:
Not sure what you're asking. The GFC700 has programmed airspeed limits, and if you exceed them, it turns off. For a 182 with G1000, it's 70-165 knots.
Click to expand...
If I understood you correctly, you mentioned that even with a wing leveler, you need to manage pitch with an AI or altimeter interface. Why? The essence of partial panel is needle, ball and airpeed. The wing leveler takes care of needle, the human manages ball with feet and pitch with airspeed.
 
It's been pretty widespread IMC the entire week up here, continuing into today. I doubt there were any pockets even remotely nearby - at least a 1 hour trip, most likely to the west or northwest, would probably have been the closest.

As for another comment about the AHRS in the Stratus 2, my experience is exactly the same - suction cupped to a side window in a 172SP, it is never accurate. Zero it out, even in straight and level, and within 15 seconds it is showing a turn and a climb or descent. They claim it should work in any orientation, but clearly mounted vertically, it does not. (for me and the other poster at least).
 
Chip Sylverne said:
If I understood you correctly, you mentioned that even with a wing leveler, you need to manage pitch with an AI or altimeter interface. Why? The essence of partial panel is needle, ball and airpeed. The wing leveler takes care of needle, the human manages ball with feet and pitch with airspeed.
Click to expand...
If you want a wing leveler to take care of pitch for you, it must have a source of reliable information.

Airspeed does not qualify because I've yet to see any light aircraft with auto throttle, and the AP can't know you haven't been messing with throttle. A true wing leveler is not alt hold either, so it requires pitch information. That's gone in the current scenario.
 
MAKG1 said:
I don't see that.
Click to expand...

You don't see what?

MAKG1 said:
Even for a rated and current instrument pilot, electing to enter IMC with a partial panel when alternatives exist is questionable judgement. Yes, we're trained to fly with a dead vacuum pump, but it is always safer to remain in VMC if the option exists.
Click to expand...

Duh. Thanks, Mr. Obvious. But where is the evidence that he had an alternative? We don't know yet that he could have gotten down without going IMC. Assuming he climbed on top, and then lost his gyro, now what? He was in VFR at one point, but during the transmission he was back in IFR conditions. He jokingly asked ATC for VFR weather. So we know he was looking to avoid IMC. We don't know where he could have safely gone to in order to avoid an IMC descent.
 
Reading all this, it's dawned on me I was undertrained. While I did a lot of partial panel work I never really did much training in recognizing impending instrument failure. A round piece of paper with a scotch tape loop on the back that had "out to lunch" written on it (yeah, my instructor was a character) suddenly appearing on the dial is different from it failing in flight and maybe still giving information, but wrong information. A little time with a CFI and a sim is in my near future.
 
ChemGuy said:
thats what i was getting at. Even if rated and current, going into IMC for more than a quick thru the deck, with a know panel issue, and possibly more failures or pilot issues was a bad idea. But if that VMC weather was an hour in the wrong direction and he couldnt spend the time or money to go that way, then i would definitely say get-there-itis played a role.
Unless there was no reachable VMC or scattered/broken layers to go thru. Then you just have to tighten up, fly right and punch thru.
Click to expand...

ATC did tell him that the weather at his destination was the best weather. He then reported that he was in IMC, and asked the controller for clear skies. So, I am not sure that get-there-itis was the issue. He specifically asked where the weather conditions were best. He just lost it in the clouds.
 
Wow got her up to 233kts!

At 233, if he did a very steep turn he could have easily done 5.7Gs at a level 80degree turn, in a descending spiral, really yanking back to get altitude, that could have some major Gs.

Looks like, presuming the owner was the pilot, he was a ATP and a CFII (expired).


Depending on the cloud levels, if all else failed what's the thoughts on getting above the layer and just stalling her into VMC?
 
Last edited:
Chip Sylverne said:
Very widespread cloud deck yesterday. Even with 2+ hours of fuel, he may not have been able to get down to vfr without dropping through the deck at some point.
No one's mentioned the possibility of an iced up pitot tube. Freeze level up there today is 4k, not much different conditions than yesterday, or earlier in the week for that matter. In the confusion, forgetting to turn on pitot heat...recipe for disaster.
Click to expand...

Yep. I wondered about the possibility of a frozen pitot. That might explain the loss "more of their panel." Of course, it could be something else, too.
 
Reading about this report yesterday prompted me to fire up the sim, set up a vacuum failure and deal with it. I tried it with ForeFlight synthetic vision (since I do have a Stratus in the airplane I fly) for a few minutes, then I turned off ForeFlight and continued. I soon realized the aircraft in the sim had an electrically powered DG, so I still had a working HSI. Even with that, the workload was really high. The complexity of dealing with the configuration changes on approach made for a tough time (I forgot how much drag the gear in the Baron has in the sim), but I got it down. It was stressful, though, and a real eye opener.

Even knowing the failure was coming (a luxury you'd never have in real life), I still had trouble ignoring the AI for the first few minutes of the scenario, especially when the AI was only slightly off compared to the other gauges. X-Plane does a good job of winding down the gyros slowly. The AI didn't have any flags, so the only clue that something was wrong was the vacuum gauge (thank god I have a 'low vac' light in the Lancair) and the fact that the AI was disagreeing with the evidence being presented by the other instruments.

This is one of those things that's really hard to simulate in the airplane, instructors would do well to use a well-equipped sim so students can see what it's really like to have to detect the problem themselves. If an instructor slaps a cover over the AI and DG and deviously exclaims, "you just lost your vacuum," that's not going to cut it. The transition from normal ops to partial panel is nowhere near that clean for most.

The things that really stood out were:
1) having a GPS showing TRK is great

2) anything that can give you a reliable AI is even better (ie, ForeFlight SV or a backup AI)

3) it takes time to transition to partial panel mode, it's not easy to ignore an AI that is only SLIGHTLY wrong for the first few minutes.

4) normally, I'm not an advocate of memorizing speeds or power settings for approaches. However, this is the one time where not having the numbers for the simulated airplane made my workload really high. I got it down, safely, but there were parts of the approach which were sketchy and I was task saturated (albeit without any backup AI).

5) if you have a plane that induces any rolling moment with a change in power setting, be prepared for some pain. I can't say I've run across this in real airplanes, but in the sim, the torque effect is slightly overstated in xplane. If I added power, there was a subtle left rolling tendency, if I reduced it, there was a right rolling tendency. This made life hard during power changes because I now had to reset the aileron trim using the turn coordinator alone.

Edit: here's a link to a video of the sim session: https://www.twitch.tv/ksmith_pe/v/64580048, staring just prior to takeoff. Vacuum pumps fail at about 4500ft. IFR clearance was heading 100, vectors ORCUT, V27, GVO, direct, maintain 5k, expect 7k in 10, dep freq 120.55 with a discrete squawk.

If you don't have a sim and have wondered what a gyro failure might look like, this gives some indication. The only issue with this simulation in terms of the gyro is that in my experience, the AI will still tend to move around (almost comically) whereas this one drifts off an eventually stops all movement.

In case anyone picks up on it, the reason for the drop in altitude just after pausing to change a nav radio frequency was caused by a sudden wind shift at altitude. The ground speed increases by about 20kts as a result, you can see it in the synthetic vision.
 
Last edited:
SixPapaCharlie said:
As far as chute parameters, I think Cirrus has changed things a bit.
@FastEddieB or @RudyP can correct me if I am wrong.

The parameters are <133kts and above 500 or 1000 feet depending on model.
But I believe Cirrus is now saying "Those are the parameters but if you are out of options, pull"
I believe the chute has been successfully deployed > 170kts and as low as 300 feet. It is not within parameters but there are people that survived outside the parameters.

If memory serves, I got this information from a video put out by Cirrus a few months ago.
Click to expand...

You're pretty close. Here are a couple of minor corrections.
- First, Cirrus remains relatively mum on expanding the 'recommended deployment envelope', while COPA (Cirrus Owners and Pilot's Association) has made great strides in expanding the knowledge based, awareness, and yes, the appropriate envelope of when to use the CAPS system
- Cirrus' testing only went up to 133KIAS (not KTAS, mind you) so the maximum demonstrated CAPS airspeed is 133 KIAS - but this is like the max demonstrated crosswind component - it is a not a limitation of use.
- There have been several successful deployments in anger in the high 180-190 KIAS range, I believe 1 was over 200KIAS and one well known failure in a steep dive at over 300 KIAS (for reference VNE on the Cirrus is 201KIAS)
- Minimum deployment altitude is now commonly viewed in the COPA world as 500 AGL for the G1-G3 airframes with the original chute and 600 AGL for the G5's with the larger and slower opening chute. You are correct that there have been very low successful deployments (including one AFTER the pilot struck power lines so ~100ish AGL) but that's really rolling the dice.
- Every Cirrus pilot should have a clear plan, brief the plan and stick to it. CAPS is not magic and will not save a pilot who gets himself in a graveyard spiral over VNE but it IS effective in a lot of situations - like anything else in the flying world, it requires some common sense
 
Last edited:
MAKG1 said:
If you want a wing leveler to take care of pitch for you, it must have a source of reliable information.

Airspeed does not qualify because I've yet to see any light aircraft with auto throttle, and the AP can't know you haven't been messing with throttle. A true wing leveler is not alt hold either, so it requires pitch information. That's gone in the current scenario.
Click to expand...

A wing leveler like the Brittain, Century 1 and Stec have an internal electric gyro that supplies the reliable bank information independent of the aircraft's vac system. No nav or pitch info, just holds the wings level, and does a fine job of it. Every instrument pilot should know the airspeed produced by a certain power setting. Set the power, get the target airspeed. If you're faster than expected, you're pitched down. Slower, pitched up..confirm with alt and VSI. If you don't have to worry about keeping the wings level, maintaining alt, climbs and descents are much easier. Basic partial panel flying.
 
James331 said:
Wow got her up to 233kts!

At 233, if he did a very steep turn he could have easily done 5.7Gs at a level 80degree turn, in a descending spiral, really yanking back to get altitude, that could have some major Gs.

Looks like, presuming the owner was the pilot, he was a ATP and a CFII (expired).


Depending on the cloud levels, if all else failed what's the thoughts on getting above the layer and just stalling her into VMC?
Click to expand...

Probably could get away with it in a Cherokee or 152-172 fixed gear that you have to work at to get to spin. A Bo is kind of slick for that. If he started getting fast he could have dropped the gear. Or maybe he did...
That's a lot of qualifications to only last two minutes. Something else musta gone wrong...Or it wasn't the owner.
 
I was thinking of dropping the gear, holding a fixed airspeed, your turn and bank is electric, so once you're in actual just step on the high wing of the turn and bank, hold airspeed, id wager you would have a manageable decent rate and would make for a quick recovery once you broke out.
 
James331 said:
I was thinking of dropping the gear, holding a fixed airspeed, your turn and bank is electric, so once you're in actual just step on the high wing of the turn and bank, hold airspeed, id wager you would have a manageable decent rate and would make for a quick recovery once you broke out.
Click to expand...

Yep, hands off the yoke, set a power for descent, drop the gear and steer with your feet, assuming your plane is in reasonable rig would work. But then again, clngs in his area were 600' agl at the time.
 
coma24 said:
Edit: here's a link to a video of the sim session: https://www.twitch.tv/ksmith_pe/v/64580048, staring just prior to takeoff. Vacuum pumps fail at about 4500ft. IFR clearance was heading 100, vectors ORCUT, V27, GVO, direct, maintain 5k, expect 7k in 10, dep freq 120.55 with a discrete squawk.
Click to expand...

We're on the second page of this thread so hopefully threadjacking is more acceptable now. What do you use for ATC with X-Plane? I had participated in VATSIM back in about 2003, using MSFS, but when I thought about getting back into that more recently since I'm flying real airplanes now, it seemed that VATSIM controllers are about 90% in Europe. The built-in ATC is, shall we say, sub-realistic.
 
I haven't used the Stratus 2S AHRS in flight but my ground observations were disappointing and I wouldn't recommend it for IMC use in any orientation. It does OK in the very short-term but they seem to have had a lot of trouble dealing with drift and resolved it by assuming that any stable attitude is straight and level. What I mean by that is if you prop it on its side and leave it for a minute or so, the virtual AI will suddenly snap from a banked attitude to wings level. The filtering has a poor response and there's quite a bit of integrator error and drift if you roll or pitch between two points several times. Their sensor fusion algorithm is poor compared to OpenPilot and cheap consumer quadcopter toys. It is way behind the state of the art and more of gimmick than a useful feature of the Stratus.
 
Chip Sylverne said:
Yep, hands off the yoke, set a power for descent, drop the gear and steer with your feet, assuming your plane is in reasonable rig would work. But then again, clngs in his area were 600' agl at the time.
Click to expand...

I thought one of the other areas had like a 1200' OVC?

Personally I'd have just broken off IFR and maneuvered around looking for a hole, otherwise I'd just take the best wx airport with a approach with vertical guidance.
 
ollopa said:
I haven't used the Stratus 2S AHRS in flight but my ground observations were disappointing and I wouldn't recommend it for IMC use in any orientation. It does OK in the very short-term but they seem to have had a lot of trouble dealing with drift and resolved it by assuming that any stable attitude is straight and level.
Click to expand...
Could be your installation.

I have used/played with the AHRS on the Status 2 in my biplane and in the Baron. It works very well in flight for me.
 
James331 said:
I thought one of the other areas had like a 1200' OVC?
QUOTE]

Think the controller told him it was 1600 and 10 at Hartford. Appears he entered IMC enroute unfortunately.
Click to expand...
 
You must log in or register to reply here.
Back
Top