Plane crashes into homes in San Diego, Ca

I looked over this new A/P data. I figured out the ADS-B track replay doesn't play at actual speed as 1X is too slow and 5X is too fast when trying to match up to the ATC tape.

A few other things I noticed. If the data is accurate, the plane doesn't seem to maintain or capture altitude while it indicates A/P on. Then the A/P kicks off and stays off the rest of the time. The ALT SEL does show being set to 3000 feet on the initial missed clearance, then the 3800 feet and the 090 heading, but it never shows it set to the 4000 or the 5000 feet. What is strange is how the pilot sounded like everything was normal when he is asked "say altitude". He responds "2500" when he should have been climbing to 4000.

I did noticed on an earlier flight the "ALTHOLD" note on ADSBexchange was displayed even as the plane was landing. So, not sure how consistently reliable that information is, unless that button is selected on the autopilot without the autopilot actually being on. During the final vectors to this approach the ALTHOLD appears by itself at some strange times.

Wondering that myself. Maybe just the flight director with the command bars slaved to the ALT HOLD?
 
Not on the STEC 3100 (that he had). When you first engage the AP it’s in pitch and roll mode to capture what you have. You can do vertical with the push of a button or lateral or both indiscriminate of each other.
Interesting. I don't have much experience with the STEC equipment. Thanks
 
Then in the course of the vectoring/climb to the right the A/P goes off and on in various modes. Like he is trying to re-engage it, but the airplane isn't doing what he thinks it should.
I see a fair amount of that when pilots look at the mode select panel to try to figure out what it’s doing rather than the annunciations on the PFD. It can get to be a vicious cycle.
 
Can you push the ALT button without the autopilot actually engage and get some result or benefit?

If the autopilot is in the Flight Director mode, maybe?

On my Avidyne DFC90 (different from the autopilot in this plane), if I start with the autopilot engaged in a certain mode like ALT, and then if I disengage the autopilot in by pushing the Flight Director button, the autopilot computer will still do something in ALT mode: it will display the Flight Director guidance for that mode on the PFD, without causing the servos to do anything to the flight controls. (This is not a common way of disengaging the autopilot, and in fact the CFI who did my transition training didn’t understand it.)
 
Unfortunately I think he was overloaded.

One thing I was taught in my instrument training is not to automatically disconnect the autopilot when things aren't going correctly, unless it is very obvious there is something wrong with the autopilot.

It's like having an extra set of hands there, and allows you to focus on the problem at hand rather than having to take over flying the airplane, you have more bandwidth to solve problems.
I can see a certain logic to that. But I think I would consider "...unless it is very obvious there is something wrong with the autopilot..." might be taking it a little far. If it doesn't seem to quite be making sense, maybe it's time to push the button, and not continue to try and find out if something is 'obviously' wrong with it. Especially on final for an Approach. Not arguing directly with you Paul. Just food for thought.
 
I can see a certain logic to that. But I think I would consider "...unless it is very obvious there is something wrong with the autopilot..." might be taking it a little far. If it doesn't seem to quite be making sense, maybe it's time to push the button, and not continue to try and find out if something is 'obviously' wrong with it. Especially on final for an Approach. Not arguing directly with you Paul. Just food for thought.

I basically agree with you, especially on final, but maybe the correct answer is go missed and figure it out rather than hand fly down the approach. There are so many different things that can occur, and for some of those disconnecting the AP would be absolutely appropriate. I guess my point is automatically disconnecting the autopilot when things aren't going correctly should not replace reasoned deliberate, thought out actions to solve a problem.
 
I basically agree with you, especially on final, but maybe the correct answer is go missed and figure it out rather than hand fly down the approach. There are so many different things that can occur, and for some of those disconnecting the AP would be absolutely appropriate. I guess my point is automatically disconnecting the autopilot when things aren't going correctly should not replace reasoned deliberate, thought out actions to solve a problem.
Gotcha. Agree. More food for thought. With an RNAV navigator/FMS or whatever 'box' and a connected Autopilot, pushing the TOGA, or equivalent button, that is a valid logic. Thing with this scenario though is it never reached the point of 'Missed Approach.' The Controller just began to Vector and Climb him. A Pilot could decide to go Missed Approach and sort out that he wasn't complying with his last heading and altitude assignment later. There are clues here that reconciling the differences between what was loaded into the 'box' via 'Approach Activated' and the instructions ATC were giving him may have contributed to the outcome.
 
I still consider myself a relatively new IFR pilot, even though it's been a few years since the checkride. I took longer than average to get my rating, and while it ticked me off that I did take longer, I experienced a lot of things in actual that someone who works to get their ticket in record time may not. The most important lesson I was taught, was that when something does not go to plan, you can't dwell on it, meaning you can't let it upset you inflight, you fix it then shake it off. The other thing that I was taught is to stay ahead of the airplane, and to go missed if you are so seriously behind the airplane you can't catch back up.

I was taught that a circle to land approach is a more risky approach for many reasons, that some commercial operations don't allow them, and other commercial operations require higher than published mins to execute them.

To those who are not familiar, for a circle to land approach, you fly a published approach to another runway, usually to higher minimums, when you break out, if you see the airport, you circle to the desired runway and land. The minimums for the circling approach is usually lower than pattern altitude, you are required to maintain visual contact with the runway, if you lose contact, you execute a missed, which involves maneuvering to get back aligned with the original runway and going. It's a more involved maneuver, especially at mins.

This approach sounds like it was easy, the ceilings were high, you just had to break out, then fly a normal pattern. This poor guy had some type of issue.

Adding another thought about ctl. I’m slowly working my way through IR training. One of my CFIIs shared how a student was failed on a checkride. The normal missed approach to the runway is a right turn west due to mountains to the east. The DPE let the pilot complete the circle then when on final told him to go missed. The pilot followed the missed right turn instruction, east towards the mountains. Automatic fail.
 
I can see a certain logic to that. But I think I would consider "...unless it is very obvious there is something wrong with the autopilot..." might be taking it a little far. If it doesn't seem to quite be making sense, maybe it's time to push the button, and not continue to try and find out if something is 'obviously' wrong with it. Especially on final for an Approach. Not arguing directly with you Paul. Just food for thought.

I basically agree with you, especially on final, but maybe the correct answer is go missed and figure it out rather than hand fly down the approach. There are so many different things that can occur, and for some of those disconnecting the AP would be absolutely appropriate. I guess my point is automatically disconnecting the autopilot when things aren't going correctly should not replace reasoned deliberate, thought out actions to solve a problem.

I'm an advocate of pushing the big red button if George decides to go nuts. Get the airplane under control, THEN figure it out.
 
...I once went missed on an RNAV approach...I went missed...he let me try again without an unsat. Passed my instrument check ride..,
The identical thing happened to me on my IFR check ride. I asked the DPE for "delay vectors to address an automation issue". He said he liked the fact I kept my cool and flew the plane. It's easy to keep your cool when you have an experienced pilot sitting next to you. I hope I'm never in the situation of solo, in the clouds and contending with an emergency. I'm sure it's a completely different animal.
 
Adding another thought about ctl. I’m slowly working my way through IR training. One of my CFIIs shared how a student was failed on a checkride. The normal missed approach to the runway is a right turn west due to mountains to the east. The DPE let the pilot complete the circle then when on final told him to go missed. The pilot followed the missed right turn instruction, east towards the mountains. Automatic fail.
Make sure you understand what the AIM says about missed approaches during the circling maneuver. :cool:
 
Gotcha. Agree. More food for thought. With an RNAV navigator/FMS or whatever 'box' and a connected Autopilot, pushing the TOGA, or equivalent button, that is a valid logic. Thing with this scenario though is it never reached the point of 'Missed Approach.' The Controller just began to Vector and Climb him. A Pilot could decide to go Missed Approach and sort out that he wasn't complying with his last heading and altitude assignment later. There are clues here that reconciling the differences between what was loaded into the 'box' via 'Approach Activated' and the instructions ATC were giving him may have contributed to the outcome.

I disagree that this was not a missed approach. I mean it's semantics I guess, but he missed the approach, and the controller cancelled his clearance. The controller then assigned a heading and altitude. Flying with the automation he had, I would have established the climb, dialed in the heading, gone to heading mode, set the altitude bug, set the fms for a climb, made sure what I entered is happening, turned the AP back on and let the AP fly, actually for me, the AP would have been on already. Then I would take a breath, and try to figure out what just happened.

I would say I'm maybe 50/50 flying missed in training situations, meaning about half the time, if I don't ask, they want the published missed, the other half I get vectors. It's just part of flying IFR, you should be completely comfortable with both.

Again, it appears this guy was confused, turning off the AP because you are confused is a bad practice. I think the problem is, for what ever reason, when he left the localizer, he thought he was doing the proper thing, then had his world blown up when the controller told him he was off course and cancelled his clearance.
 
I'm an advocate of pushing the big red button if George decides to go nuts. Get the airplane under control, THEN figure it out.

Depends on what "goes nuts" is. I had an autopilot do an uncommanded climb before, that made me disconnect. But again, if I'm confused, but the air plane is flying fine, I'm leaving George on and working on getting unconfused.
 
Another possibility was that he had broken out at the time he inquired if he was cleared to land on 23. He may have started the circle which is why he was diverging from the localizer course. Then peppered with a pile of confusing to him instructions, he claimed back into the clouds without the benefit of having George engaged. Still, it seems bizarre you can't knuckle down and get into a climb when the controller is issuing countless low alt warnings. He'd have likely been back on top pretty soon.
 
Another possibility was that he had broken out at the time he inquired if he was cleared to land on 23. He may have started the circle which is why he was diverging from the localizer course...

Let's say he did start the maneuver then. He would be one lucky pilot to suddenly see the airport and runway just when it's time to intercept the localizer. But it would also show he wasn't familiar with the circling maneuver. Can't say I've heard any guidance on starting a circling maneuver from 7 miles out. I'm thinking its more probable he did have some ground contact and may have been doing his own navigating at that point. Or, he was already overloaded and lost his SA fast.
 
Blancolirio's report is excellent. Sorry, I'm an infrequent user, don't meet criteria to post links. If unfamiliar, go to youtube/user/blancolirio and click on the videos button. This event is the latest analysis.

The pilot wasn't making a circling approach when he crashed. He was on an IFR flight plan to KMYF, cleared to intercept the localizer (not yet cleared for the approach itself). He missed the localizer and turned to the north in IMC. He increased engine power and accelerated. ATC repeatedly gave him vectors and instructions to climb, with increasing urgency as he approached rising terrain. Still IMC, the pilot reported climbing when he actually was descending (and turning right). Tone of voice suggested high stress level. Juan interprets all this as clear indications of somatographic illusion, with acceleration in IMC interpreted as a climb, and reacting to body sensations and not his instruments. It is highly probable he had increasing spatial disorientation and finally a death spiral to the right until ground contact at high speed just north of KSEE.

All of which raises a question I have long had about instrument rated pilots based in places where they infrequently make real-world approaches in IMC, like Arizona. The pilot lived in Yuma, where the flight originated. I wonder how many approaches he had made, and how often, in IMC. (Sorry if I missed a discussion of this issue among the many posts above.)

Hunter Handsfield
T182T N134GW
BFI, Seattle
 
Last edited:
I disagree that this was not a missed approach. I mean it's semantics I guess, but he missed the approach, and the controller cancelled his clearance. The controller then assigned a heading and altitude. Flying with the automation he had, I would have established the climb, dialed in the heading, gone to heading mode, set the altitude bug, set the fms for a climb, made sure what I entered is happening, turned the AP back on and let the AP fly, actually for me, the AP would have been on already. Then I would take a breath, and try to figure out what just happened.

I would say I'm maybe 50/50 flying missed in training situations, meaning about half the time, if I don't ask, they want the published missed, the other half I get vectors. It's just part of flying IFR, you should be completely comfortable with both.

Again, it appears this guy was confused, turning off the AP because you are confused is a bad practice. I think the problem is, for what ever reason, when he left the localizer, he thought he was doing the proper thing, then had his world blown up when the controller told him he was off course and cancelled his clearance.
It's semantics. While I wouldn't call that a "missed approach" I get that it might be called that. I get what you say about not automatically turning off the Autopilot just because you are confused about something. But if the confusion is about what the Autopilot may or may not be doing, I'd probably 'push the button.'
 
Last edited:
Been here a few times.

Let's say doc had the minimum 1MM/100k sublimit liability coverage.

All of the claimants will assemble their damages. There will be dozens of 'injured's. Every attorney will be claiming for the full amount (1MM)

The insurance company will attempt to settle within the policy limits. Nobody will take that deal, so no progress can be made -- insurance needs to settle everything or nothing.

The insurance company now has $1MM to spend defending the doc against all of the claims against him, which is a decent amount of lawyer-power. The ones with a body count or property damage will prevail and get something. All of those karens who sued for PTSD will hopefully get told to kick rocks (our karens did -- doc may not be so lucky since he likely has an estate of some size, the greed will be enhanced)

It will take a year or two to unwind it all.

I had one case where a biz jet crashed into a few homes, killing some passengers, injuring others, and demolishing the homes, one of which had someone inside. We invited everyone with a claim to a mediation, where we said we would pay the limits out if they agreed amongst themselves how to divide the policy. We did have the one lady in one of the houses who didn't come. She was not injured physically, and was just claiming emotional distress. (She alleges she was kneeling down in her living room praying at the exact moment the plane crashed into her home.) So we carved out a small amount for her claim, settled with everyone else (with my client's, the insured's, consent), and took the risk of an excess judgement from her to put the rest of the claims to bed. We ended up litigating her case for some time before she finally took the small bit of money we had left on the policy to settle her claim.
 
Last edited:
AOPA has a preliminary discussion and what the NTSB will most likely be looking at
 
Last edited:
Juan, as usual, has a great review. I like that he describes the potential pilot incapacitation, as spacial disorientation.

 
Juan, as usual, has a great review. I like that he describes the potential pilot incapacitation, as spacial disorientation.

Juan Browne is a hack. I watched his video this morning. He has some good points/details in there, but as is usual for Juan, he goes into his analysis with great confirmation bias and overlooks several key details in the process.

Best videos I've seen on this accident so far are the VASAviation one and the above linked Air Safety Institute.
 
A loss-of-control accident may have had one final opportunity to avoid disaster, before or after exiting IMC, nose down with excessive bank angle
... recovery from unusual attitude.

Formerly for instrument currency, (if using a sim/FTD/ATD), CFR 61.57(c) included:
-Two unusual attitude recoveries while in a descending, Vne airspeed condition
and
-two unusual attitude recoveries while in an ascending, stall speed condition.

Although no longer required in regs, 'recovery from unusual attitudes' (practice overcoming vertigo/disorientation) ... might have led to a different outcome.
Thoughts and condolences for family and friends of pilot and ground victims; a tragic loss.
 
Last edited:
A loss-of-control accident may have had one final opportunity to avoid disaster, before or after exiting IMC, nose down with excessive bank angle
... recovery from unusual attitude.

Formerly for instrument currency, (if using a sim/FTD/ATD), CFR 61.57(c) included:
-Two unusual attitude recoveries while in a descending, Vne airspeed condition
and
-two unusual attitude recoveries while in an ascending, stall speed condition.

Although no longer required in regs, 'recovery from unusual attitudes' (practice overcoming vertigo/disorientation) ... might have led to a different outcome.
Thoughts and condolences for family and friends of pilot and ground victims; a tragic loss.
In my experience, the majority of pilots execute the proper unusual attitude recovery procedure only after hearing the words “put your head down and close your eyes.”
 
In my experience, the majority of pilots execute the proper unusual attitude recovery procedure only after hearing the words “put your head down and close your eyes.”

In my recent flight review the instructor had a twist on this as she had me keep the controls, close my eyes, and then make a descending left turn. After a half minute or so I was certain that I must be screaming towards the ground at a fairly steep rate. Imagine my surprise to open my eyes and see that we were pretty much wings level and climbing at a pretty good nose up attitude!

We did most of the regular flight review stuff but her method of unusual attitudes was a bit different.
 
In my recent flight review the instructor had a twist on this as she had me keep the controls, close my eyes, and then make a descending left turn. After a half minute or so I was certain that I must be screaming towards the ground at a fairly steep rate. Imagine my surprise to open my eyes and see that we were pretty much wings level and climbing at a pretty good nose up attitude!
My instrument DPE did something similar, gave me a bunch of climb/turn/descend instructions with my eyes closed. After a minute or so, he said "This isn't working. Are you peeking? My airplane."
 
In my recent flight review the instructor had a twist on this as she had me keep the controls, close my eyes, and then make a descending left turn. After a half minute or so I was certain that I must be screaming towards the ground at a fairly steep rate. Imagine my surprise to open my eyes and see that we were pretty much wings level and climbing at a pretty good nose up attitude!

We did most of the regular flight review stuff but her method of unusual attitudes was a bit different.
I’ve done that as well…you’re still putting it in their heads that it’s time to think about unusual attitude recovery. And there are some who refuse to cooperate for whatever reason.

I’ve got a sim malfunction that results in a nose-high unusual attitude, and the only way to survive is the proper nose high recovery technique. Sometimes I even verbally review the recovery procedure beforehand, and most pilots still “die”.
 
My instrument DPE did something similar, gave me a bunch of climb/turn/descend instructions with my eyes closed. After a minute or so, he said "This isn't working. Are you peeking? My airplane."

The desire to peek is very strong! But I realized that doing so wouldn't help me learn anything at all so I kept them closed.
 
The desire to peek is very strong! But I realized that doing so wouldn't help me learn anything at all so I kept them closed.
Yes it is, but I didn't peek. I really did fly all of that with my eyes closed.

I should also add that after the DPE took over, my nose-high recovery was a little slow, so it's not like I'm a super-pilot or anything.
 
Juan Browne is a hack. I watched his video this morning. He has some good points/details in there, but as is usual for Juan, he goes into his analysis with great confirmation bias and overlooks several key details in the process.
Professional speculation seems to have grown into a legitimate sideline for some.
 
Professional speculation seems to have grown into a legitimate sideline for some.
People are finding ways to make money off of youtube.

Just like with the media, there is a wide range of quality and style.

Dan Gryder is pretty much the National Enquirer - focus on sensational juicy tabloid details
Juan Browne is CNN - facts often get lost in bias. He editorializes and calls it 'Analysis'
Scott-Perdue (Fly-Wire) is more like the Wall Street Journal - good analysis, far less bias

And Flight Chops is Guy Fieri
 
Dan G has chalked this up to an adverse reaction to a C19 booster shot. We are living inside of a Saturday Night Live sketch.
 
DJkWzaP.gif

:cornut:
 
I though we were supposed to keep politics out of this?

Seems he answered his own question (but that didn't stop him).

"I’m not a medical doctor, so perhaps I’m applying my ignorance to questions I ought not to attempt to pose ... "

Yeah it's time ... IBTL
 
Back
Top