Landed on the water. Why?

[marcosvc]

I am posting this to help my friend in Brazil. Yesterday he lost his Mini2. It was 80% battery, was returning, very close to the home point and without any command, started to go down and bye bye....
The log is attached. There is a messagem of Motor blocked (30165) but the drone went down very slowly.
Could anyone help me about the cause or possibilities that had occur?
I would appeciate a lot. Thanks in advance.
Follow the video about de crash

View this content on Instagram

View this content on Instagram

 

[Yorkshire_Pud]

The log is attached. There is a messagem of Motor blocked (30165) but the drone went down very slowly.

No log.
Upload it to

DJI Flight Log Viewer | Phantom Help

Upload and view flight logs from your DJI drones

www.phantomhelp.com

then post the URL here.

 

[marcosvc]

DJI Flight Log Viewer - PhantomHelp.com

 

[marcosvc]

Update. After checking the log, i thing he made a mistake and went in the water by himself. The left stick was down in the last 5 seconds.... so probably this is the crash reason.

 

[Yorkshire_Pud]

Update. After checking the log, i thing he made a mistake and went in the water by himself. The left stick was down in the last 5 seconds.... so probably this is the crash reason

Unfortunate.

 

[Capt KO]

Update. After checking the log, i thing he made a mistake and went in the water by himself. The left stick was down in the last 5 seconds.... so probably this is the crash reason.

Pilot error..in the dark confusion. Or stuck stick

 

[marcosvc]

Pilot error..in the dark confusion. Or stuck stick

I think the great possibility to make this error was the transmission delay. It was very hot 40 Celcius, the mini2 does not have cooler. He was receiving signal messages all time. Because of the transmission delay, he supposed was high but the drone was further ahead. At night, all the factors contribuited for the crash.

 

[Yorkshire_Pud]

I think the great possibility to make this error was the transmission delay.

Personally I think that is unlikely, sorry.
Why would temperature influence lag?
As far as I know the only temperature recorded in the drone is the battery's and that reached 49.4C. I have had several mini or mini 2 batteries in a mini 2 reach 70C+ .
Although 40C is DJI's stated maximum air temperature I don't think there are any temperature warnings in the log.

 

[marcosvc]

Personally I think that is unlikely, sorry.
Why would temperature influence lag?
As far as I know the only temperature recorded in the drone is the battery's and that reached 49.4C. I have had several mini or mini 2 batteries in a mini 2 reach 70C+ .
Although 40C is DJI's stated maximum air temperature I don't think there are any temperature warnings in the log.

We live in a very hot country. In that day the temperature was 40C official (inside an enclosure place) but outdoor it reached 50C. I dont know how this affects the drone's communication, but I see a lot of reports of that in hot days. The log reported lost link, weak signal, loss of remote control, etc. These signal problems generates the signal delay that I mentioned. Also, talking to the drone's operator, he reported that he saw the drone flying oon the screen after it droped in the water.

 

[Yorkshire_Pud]

If you saw 50C as an air temperature then you need to contact your authorities. Quite a few websites are posting reports of the recent temperatures setting a new record of 44.8C in a place reported as Araçuaí.
Besides aren't temperatures generally cooler on the coast and at night?

If he saw it fly after the dunking then why didn't he fly it back home?
I think your friend made a mistake and that's the whole story.
They are not the first to make a mistake that cost the drone and they will not be the last.

 

[Meta4]

I think the great possibility to make this error was the transmission delay. It was very hot 40 Celcius, the mini2 does not have cooler. He was receiving signal messages all time. Because of the transmission delay, he supposed was high but the drone was further ahead. At night, all the factors contribuited for the crash.

The mini 2 has no problem flying in 40° temperatures and as long as it's flying, will be adequately cooled.
40° temperature does not cause any transmission delay.

He was flying 14.8 metres above the water.
The reason the drone crashed into the water is 100% due to the flyer pulling the left stick fully down at 5:45.4 and holding it down until the drone crashed into the water 7 seconds later at 5:51.4.

The drone didn't "land" in the water because as well as holding the left stick down, he also pushed the right stick forwards at 5:47.5 and held it there causing making the drone accelerate.
It had reached a speed of 8 metres/sec when it crashed into the water.

It's most likely that the flyer was disoriented by flying at night and not having clear vision of the drone's position relative to the water surface.

 

[Mynah]

I think the great possibility to make this error was the transmission delay. It was very hot 40 Celcius, the mini2 does not have cooler. He was receiving signal messages all time. Because of the transmission delay, he supposed was high but the drone was further ahead. At night, all the factors contribuited for the crash.

Very possible, I had a close encounter flying a Mini 4 Pro on a tablet that was very happy flying the Mini 2 and Mini 3 and Mini 4 Pro. But flying the Mini 4 Pro AND recording screen resulted in screen lag of about 5 seconds, I realized this when noticing the delay of control stick input on the way back. I had to land ignoring the screen video feed.

The scary part was that the lag only started at the end of the mission so was not consistent. If this happened during intricate flying maneuvers trusting the video feed the story might have ended sadly.

So yes, if he developed screen lag on his way back in the dark flying in a straight line he would not have noticed it. Him stating that he still saw the drone flying on his screen AFTER it went down explains it.

As for the reason of screen lag one can only guess. In my case the cause was clear, a tablet GPU that simply could not cope with screen recorder and the demand Fly and the Mini 4 Pro put on it. No problem screen recording and flying the other Minis. But the tablet did not crash the Fly app, only lagged, crashing the app might even be safer than unknowingly flying with a 5 second screen lag.

The tablet in question was a 2023 release Ulefone Armor Tab on Android 12. But DJI Fly 1.12.3 and the specific Mediatek GPU are not friends, I have tried every trick in the book to try make it work.

 

[MS Coast]

Very possible, I had a close encounter flying a Mini 4 Pro on a tablet that was very happy flying the Mini 2 and Mini 3 and Mini 4 Pro. But flying the Mini 4 Pro AND recording screen resulted in screen lag of about 5 seconds, I realized this when noticing the delay of control stick input on the way back. I had to land ignoring the screen video feed.

The scary part was that the lag only started at the end of the mission so was not consistent. If this happened during intricate flying maneuvers trusting the video feed the story might have ended sadly.

So yes, if he developed screen lag on his way back in the dark flying in a straight line he would not have noticed it. Him stating that he still saw the drone flying on his screen AFTER it went down explains it.

As for the reason of screen lag one can only guess. In my case the cause was clear, a tablet GPU that simply could not cope with screen recorder and the demand Fly and the Mini 4 Pro put on it. No problem screen recording and flying the other Minis. But the tablet did not crash the Fly app, only lagged, crashing the app might even be safer than unknowingly flying with a 5 second screen lag.

The tablet in question was a 2023 release Ulefone Armor Tab on Android 12. But DJI Fly 1.12.3 and the specific Mediatek GPU are not friends, I have tried every trick in the book to try make it work.

Screen lag, if it was present in the case the OP described, doesn't explain the left stick being pulled back and held back while the drone descended from 15+ meters to the water's surface.

 

[Mynah]

Screen lag, if it was present in the case the OP described, doesn't explain the left stick being pulled back and held back while the drone descended from 15+ meters to the water's surface.

Drone controls remain in real time, only video feed lags, you'll see a wall 100m away, keep giving forward input while in fact the drone has already crashed into the wall 5 seconds earlier. He saw the ocean 15m below on his video feed, pulled back left stick and on the video feed it looked like the drone was not responding while in fact it was already in the ocean.

 

[Mynah]

Which is why I'm saying, an app crash would be safer than an app video feed lag. My first instinct during an app crash is to let go of everything I am busy with, then hit RTH knowing at what altitude it will be coming back so I can lay eyes on it and land it ignoring the luxury of having a video feed.
My first thought experiencing video lag was that the drone was not responding to my input, only to appear doing what I asked 5 seconds later. It is a mind warp at first, having flown hundreds of flights while believing what I was seeing was real.

 

[MS Coast]

Drone controls remain in real time, only video feed lags, you'll see a wall 100m away, keep giving forward input while in fact the drone has already crashed into the wall 5 seconds earlier. He saw the ocean 15m below on his video feed, pulled back left stick and on the video feed it looked like the drone was not responding while in fact it was already in the ocean.

Yes, I agree. But the OP said the drone was "very close to the home point," so presumably the pilot had the drone in sight and mistakenly held the left stick back as it descended.

 

[Yorkshire_Pud]

I have flown a mini 2 at night and am wondering why anyone would fly it eyes on the screen? If the pilot was flying via the screen then that was a mistake in my opinion.
The only time I used the screen was during the descent to landing to ensure that it was coming down in the correct place.
The drone was below a reported height of 40 ft, around 56ft AGL, from 318.5 seconds onwards, to 351.5 seconds, i.e. the last 33 seconds. I can imagine no reason to descend lower until the drone was literally and eyes-on-the-drone, over the home point - especially when the drone is over water. The final descent started at 345.3 seconds.
So, even if there was lag within the screen device, it's pilot error in my book.

 

[marcosvc]

If you saw 50C as an air temperature then you need to contact your authorities. Quite a few websites are posting reports of the recent temperatures setting a new record of 44.8C in a place reported as Araçuaí.
Besides aren't temperatures generally cooler on the coast and at night?

If he saw it fly after the dunking then why didn't he fly it back home?
I think your friend made a mistake and that's the whole story.
They are not the first to make a mistake that cost the drone and they will not be the last.

Termômetro marca 50º C em Santos

Registro foi feito na tarde deste sábado (18), em marcador digital localizado no Canal 1

www.diariodolitoral.com.br

https://costanorte.com.br/cidades/s...-de-quase-50-graus-neste-domingo-1413995.html
Some evidence that we had 50C outside in that day. This is not the official temperature measured inside, without sun. This is the outside, measured on the street.
When we have delay piloting a drone, the command are prompt, but the image takes some seconds to show in the screen. Thats why the operator saw the drone in the air, and in reality he had droped it in the sea already.
This is operator mistake because when you are close to homepoint you should pilot looking at the drone, not to the screen.

 

[marcosvc]

Screen lag, if it was present in the case the OP described, doesn't explain the left stick being pulled back and held back while the drone descended from 15+ meters to the water's surface.

See an example of vídeo lag. Isso in Portuguese. But the left image is the real one, and the rught onde is the screen app

 

[Mynah]

See an example of vídeo lag. Isso in Portuguese. But the left image is the real one, and the rught onde is the screen app

At least the screen lag in this video is jumpy so one can suspect lagging video. In my case with the Ulefone Armor Pad the video feed was smooth, only delayed by about 5 seconds. I have replaced the Ulefone Armor Tab with the new Samsung Tab A9 running Android 13, an 8 inch tablet which is relatively affordable and it works a charm.