Mavic Pro started to descend by itself above water

[thegroovologist]

I did a couple of test flights above ground...absolutely rock-solid.

When I went to a lake today and started a flight, I stopped the drone above water (10 metres away

from me), around 2 metres in the air. The Mavic Pro started to descend slowly by itself then.

I reacted immediately (gained altitude) and flew back...didn't try it again because confidence was

gone. This happened during two test flights above water (two different locations). Does anyone know

what the issue might be?

 

[msinger]

Upload your TXT flight log here and see if it explains what happened. If you need help, post a link back here after you upload your log.

 

[Cyberpower678]

Upload your TXT flight log here and see if it explains what happened. If you need help, post a link back here after you upload your log.

I think we've seen this enough times to say it's because the sensors on the bottom don't like water.

I did a couple of test flights above ground...absolutely rock-solid.

When I went to a lake today and started a flight, I stopped the drone above water (10 metres away

from me), around 2 metres in the air. The Mavic Pro started to descend slowly by itself then.

I reacted immediately (gained altitude) and flew back...didn't try it again because confidence was

gone. This happened during two test flights above water (two different locations). Does anyone know

what the issue might be?

I've seen so many threads on this forum regarding the same thing. Because the water is fluid, the downward sensors can't gauge the ground well and the drone slowly sinks above water when hovering close to it. If you climb to at least 5 meters, then you don't have this problem. Alternatively you can shut off the bottom sensors and the drone will rely on the IMU to maintain altitude and it will hold steady above water.

 

[msinger]

I think we've seen this enough times to say it's because the sensors on the bottom don't like water.

Sure, we can make some assumptions if that's all the OP is after. Viewing the flight log is the only way to know for sure though.

 

[thegroovologist]

I think we've seen this enough times to say it's because the sensors on the bottom don't like water.


I've seen so many threads on this forum regarding the same thing. Because the water is fluid, the downward sensors can't gauge the ground well and the drone slowly sinks above water when hovering close to it. If you climb to at least 5 meters, then you don't have this problem. Alternatively you can shut off the bottom sensors and the drone will rely on the IMU to maintain altitude and it will hold steady above water.

Thanks a lot for your reply! I will shut off the bottom sensors during the next test flight above water.

I wonder though what would have happened if I hadn't reacted. Can water fool the sensors so much

that the drone keeps descending by itself until it crashes?

 

[Cyberpower678]

Thanks a lot for your reply! I will shut off the bottom sensors during the next test flight above water.

I wonder though what would have happened if I hadn't reacted. Can water fool the sensors so much

that the drone keeps descending by itself until it crashes?

I'd rather not find out. But it happens to my drone too.

 

[Cyberpower678]

Thanks a lot for your reply! I will shut off the bottom sensors during the next test flight above water.

I wonder though what would have happened if I hadn't reacted. Can water fool the sensors so much

that the drone keeps descending by itself until it crashes?

@msinger will be more than happy to view your flight log to confirm.

 

[iamSteve]

I have not tried this yet, so would love someone who knows to look at your log file.... One thing I can't understand, why does the gps not hold the height?

 

[Kilrah]

Close to the ground the sonar is much more precise, so this becomes the primary source of altitude reading. You couldn't land smoothly relying only on GPS.

 

[boone]

I fly over water pretty much every time I fly. That's a circumstance of my surroundings, so I can't really NOT fly over water. And often times that water has some churning waves coming in. And I can absolutely confirm the phenomenon your claiming you witnessed. I can also say that over churning surf, the mavic will som times latch onto a wave and drift this direction or that. Not so freaky when it's an incoming wave, but heart stopping when it's backwash. I keep the sensors on, but I also keep a vigilant eye on the altitude as well. I also don't fly beyond los so I can keep an eye on it that way too. I do fly low sometimes still, and I accept that I'm doing so at my own risk. I think that the sensors on the bottom do get a little thrown off by reflections over water, but that's just an assumption. Also, I often times fly below the altitude of my takeoff point, which is flat out terrifying honestly. I need to have like three more eyeballs to be able to watch all of the things I'd need to keep an eye on for it to not be terrifying really.

 

[iamSteve]

So is it right to assume 5m+ the ultrasonics will not take over when over water?

 

[Cyberpower678]

So is it right to assume 5m+ the ultrasonics will not take over when over water?

Correct

 

[thegroovologist]

Viewing the flight log is the only way to know for sure though.

I just tried to download the flight log (Black Box) but gave up after waiting for like 15 minutes...stuck at "Exporting". It can't take that long to export flight logs, can it ?

 

[Cyberpower678]

I just tried to download the flight log (Black Box) but gave up after waiting for like 15 minutes...stuck at "Exporting". It can't take that long to export flight logs, can it ?

You can get it from the Flight Data tab. Data can be broken up into multiple files. They're broken into groups of 400MB files.

 

[msinger]

So is it right to assume 5m+ the ultrasonics will not take over when over water?

Per the Mavic manual, the downward sensors are effective when the Mavic is at altitudes between 0.3 to 13 meters.

 

[Cyberpower678]

Per the Mavic manual, the downward sensors are effective when the Mavic is at altitudes between 0.3 to 13 meters.

Oops. I thought they were only effective 7m. Thanks for mentioning that.

 

[thegroovologist]

You can get it from the Flight Data tab. Data can be broken up into multiple files. They're broken into groups of 400MB files.

When I click on Flight Data there's an "Open Data Viewer" button.

When I click on "Open Data Viewer" I see this (not sure how to proceed):

When I click on the load icon, I can see a "DJI Flight Log" folder which contains .DAT files.

The last .DAT file is "FLY047" . The video file (today's test flight) is named DJI_0060.MOV though.

 

[Cyberpower678]

When I click on Flight Data there's an "Open Data Viewer" button.

When I click on "Open Data Viewer" I see this (not sure how to proceed):

When I click on the load icon, I can see a "DJI Flight Log" folder which contains .DAT files.

The last .DAT file is "FLY047" . The video file (today's test flight) is named DJI_0060.MOV though.

It's the DAT file I want.

 

[bucefalo]

same happens in low light? like by night? the 1st flight I did by night(more than flight was take off and hover lol) and I could see same attitude like you describe on the water.. while during the day everything just perfect! can you confirm?

 

[boone]

same happens in low light? like by night? the 1st flight I did by night(more than flight was take off and hover lol) and I could see same attitude like you describe on the water.. while during the day everything just perfect! can you confirm?

I don't get altitude problems when I fly in low light, but I do definitely get the same kind of drift. And the reasons for that have been well explained by the good folks here and DJI themselves so nothing shocking.