MA2 (RC Disconnected) No RTH, Drone gone

[boblui]

There were no disconnections, either between the aircraft and the RC or between the RC and the mobile device. What "false indication" are you referring to?

I am thinking of some software glitch causing the app to show the disconnection message which was not logged. There was also the beeping which can't be explained by the log.

 

[PhantomFandom]

The log doesn't record the video feed status, but the log is recording the on-screen telemetry - that's what OSD means.

Yes I totally understand and agree with what you are saying. I am just trying to reconcile the pilot's account with what we see in the log.

 

[svenmarbles]

Yes I totally understand and agree with what you are saying. I am just trying to reconcile the pilot's account with what we see in the log.

I’m going to offer that perhaps the OP’s account of things might be incorrect.

 

[sar104]

I am thinking of some software glitch causing the app to show the disconnection message which was not logged. There was also the beeping which can't be explained by the log.

That was almost certainly after it landed in the water.

 

[sar104]

I’m going to offer that perhaps the OP’s account of things might be incorrect.

Yes - his recollection is inconsistent with the log data. The consistent explanation is that the last altitude that the pilot noticed was 93 ft, and that he then inadvertently applied down throttle for 10 seconds, after which the aircraft was down and the RC reporting disconnected.

 

[Scott]

At that point where it shows the decent is where I lost connection to the RC. I was literally looking at the screen where the drone was approximately 100ft up and coming back in and the next thing that happened was the RC disconnect. I could see my my location on the screen and then poof, signal gone. I always bring it in at least 50ft above my position then slowly take it in. I wasn't even at the point when I lost signal. I've attached my 2 logs from yesterday, showing I never throttle down at full like that when coming in for a landing. Different location but similar setting. I just don't want people to think I made some noob mistake as I've been flying for over 5 years. Just to prove, here is a video I made with the Phantom 3 Pro almost 5 years ago, when there wasn't any intelligent flight modes or obstacle avoidance sensors.

I was literally making a follow up video of different locations. All my shots are done by sticks only.

 

[sar104]

At that point where it shows the decent is where I lost connection to the RC. I was literally looking at the screen where the drone was approximately 100ft up and coming back in and the next thing that happened was the RC disconnect. I could see my my location on the screen and then poof, signal gone. I always bring it in at least 50ft above my position then slowly take it in. I wasn't even at the point when I lost signal. I've attached my 2 logs from yesterday, showing I never throttle down at full like that when coming in for a landing. Different location but similar setting.

Unfortunately that's not reconcilable with what the log recorded. As I pointed out previously, it's trivially obvious that the aircraft «» RC «» mobile device connections were up for the entire flight because the telemetry is complete all the way to splash down. It's possible that you lost the video feed, but you did not lose telemetry or control, and the RC disconnect occurred in the water at 732 seconds elapsed flight time, not at 93 ft altitude.

 

[k9kids]

So, what is the OSD_zSpeed data showing as the craft dives for the water? All other readings seem of a negative value.

 

[Dincena]

I feel your pain Scott! Recently watched my MA2 slowly decend into the east river and there was nothing i could do about it. A very expensive leason learned for me. My heart will never beat the same again flying over water after that. Sorry for your loss

 

[Meta4]

Recently watched my MA2 slowly decend into the east river and there was nothing i could do about it.

You initiated autolanding at 2:11.3 and 185 feet above the river and watched it descend for almost a minute, until it splashed down at 3:07.5.
You only tried pushing the throttle forwards but didn't cancel the autolanding which would have instantly solved your problem.

If you accidentally initiated autolanding when you thought you were initiating RTH, the result might have been similar because you were impatient to launch and did so before the drone acquired GPS and could record a homepoint.
By the time it had GPS, it was already out over the river and that's where it recorded its homepoint.
You would have had to cancel the RTH autolanding to get the drone back once that started.

 

[svenmarbles]

You initiated autolanding at 2:11.3 and 185 feet above the river and watched it descend for almost a minute, until it splashed down at 3:07.5.
You only tried pushing the throttle forwards but didn't cancel the autolanding which would have instantly solved your problem.

??‍


So of the complaints about “defective units”, how many do you suppose are examples like these?

 

[PhantomFandom]

At that point where it shows the decent is where I lost connection to the RC. I was literally looking at the screen where the drone was approximately 100ft up and coming back in and the next thing that happened was the RC disconnect.

Hey Scott, I feel your pain. As a seasoned pilot you do your best but mistakes still happen. A famous quote goes " The search for truth takes you where the evidence leads you, even if, at first, you don't want to go there.". I know you don't want to think that you flew your MA2 right into the water but right now that is where the evidence leads us.

I am still not clear what EXACTLY happened when you say that you had a RC disconnect. You were looking at the screen and then what? Very precisely, what did you see? What was on the screen? What messages popped up?

Were you shooting video at that time and were you caching to the mobile device? If so, that might be helpful to review.

Also the log is very clear that you had the left stick almost all the way down for that ten second period. There is no way to dispute that. So even if you could see nothing on the screen then why command the drone to descend rapidly like that?

 

[Meta4]

So of the complaints about “defective units”, how many do you suppose are examples like these?

Accounts like that aren't uncommon.

 

[sar104]

Accounts like that aren't uncommon.

That sums it up, I'm afraid.

 

[sar104]

So, what is the OSD_zSpeed data showing as the craft dives for the water? All other readings seem of a negative value.

The frame of reference is x (north), y (east) and z (down). So positive z velocities are downwards.

 

[Scott]

Personally I would rather have it be my fault, because now how can I trust getting another Mavic Air 2? But right now with my style of flight, (Precision, and small movements) Never sport mode, I really can't trust getting another one. Thankfully I still have my Anafi, so I can still do Real Estate Videos.

 

[Meta4]

Personally I would rather have it be my fault, because now how can I trust getting another Mavic Air 2?
But right now with my style of flight, (Precision, and small movements) Never sport mode, I really can't trust getting another one.

To put bluntly what the data whisperers above have been politely explaining ...
The person who pulled the throttle down hard caused the drone to descend into the water.
It was your fault!

 

[Scott]

To put bluntly what the data whisperers above have been politely explaining ...
The person who pulled the throttle down hard caused the drone to descend into the water.
It was your fault!

I've definitely read the data, but all I can say is I was watching live on the screen, and while obviously depth perception is limited with a phone, height perception definitely isn't. I'm pretty sure I know the difference of even 20ft vs 100ft. Plus the speed of the supposed decent in 7 seconds, when my entire return is a gradual descent. My only fault which will never happen again is I had changed phones and wasn't caching the video on the new phone (Changed phones for more NITS). That will definitely never happen again, and probably will screen record everything from now on as well.

 

[Meta4]

I've definitely read the data, but all I can say is I was watching live on the screen, and while obviously depth perception is limited with a phone, height perception definitely isn't. I'm pretty sure I know the difference of even 20ft vs 100ft. Plus the speed of the supposed decent in 7 seconds, when my entire return is a gradual descent. My only fault which will never happen again is I had changed phones and wasn't caching the video on the new phone (Changed phones for more NITS). That will definitely never happen again, and probably will screen record everything from now on as well.

So to be in complete denial, all you have to do now is find a way to account for the flight data.
The data clearly shows you flying along at full forward speed, 93 feet higher than the launch point and pulling the left stick down from 12:02.1 until and holding it there until the data ends at 12:11.8.
There was no loss of signal until the drone sank below the water.

Plus the speed of the supposed decent in 7 seconds, when my entire return is a gradual descent.

The speed of descent is exactly as would be expected when the left stick is held down where it was.
There's no unexplained mystery there.

 

[Fred Garvin]

...I was literally watching the video and definitely wasn't even close to the water and then that is where the signal to RC was gone...

So, you were flying by watching the screen and did not have VLOS on the aircraft....the video feed cut out and you had no way to fly the bird and blindly flew it into the water....I don’t think you would have done that if you were watching the bird...with your eyes...rather just flew it back to you and landed without incident.