Mavic 3 Stopped Responding Mid Flight

[Tim3459575]

Hi,

Flying my Mavic 3 yesterday, with a catch handle, in C mode with RC controller when it started stalling as I was trying to get it to fly home. Firmware is up to date. It would fly a bit then stop (and not respond to forward stick), then fly again and stop. I thought it could be wind so I tried changing altitude and there was no improvement. Eventually, I gave up trying to control it with the stick and used the "return to home" command. It responded immediately and properly, and once it was near my position I took over manual control and it seemed normal. I had direct line of sight and it wasn't that far away so I don't think it would be poor signal. I was on a sailboat though, with a metal mast and metal rigging. Attached is the flight log. Perhaps this was due to being in C mode ? Hardware issue ? Any advice on what may have caused this and what I should do to prevent it in future, would be much appreciated !

Thank you,
Tim

 

[FLDave]

By stalling yo mean it stopped moving, not really stalling or losing lift, right?
Do you have a picture of this "catch handle" showing exactly how it was attached? What is the weight of the catch handle?
What was the actual wind speed?
The flight log was not attached.

 

[Tim3459575]

By stalling yo mean it stopped moving, not really stalling or losing lift, right?
Do you have a picture of this "catch handle" showing exactly how it was attached? What is the weight of the catch handle?
What was the actual wind speed?
The flight log was not attached.

Thanks Dave, The flight record is 3 mb and it says it's too large to upload. Same for the photos. Is there another way to upload those ? By stalling, I mean hovering and not going anywhere, just holding its position. The PVC catch handle is extremely light with little wind resistance as it's only 1/4" thick. Estimate for wind was 10 mph.

 

[FLDave]

The flight log file will be a .TXT file not a .DAT file. I just upload to AirData. For uploading here just search for methods.
I have not idea what PCV catch handle looks like or how it is attached.

 

[Yorkshire_Pud]

or IMO better upload it to DJI Flight Log Viewer | Phantom Help
the output csv contains more information and is more useful.

Where was the sun in relation to the drone, was OA on?

 

[RodPad]

Yorkshire_Pud

mavicpilots.com

You beat me with your post.

Tim3459575

mavicpilots.com

Yep best direction to go!

Rod ..

 

[Tim3459575]

The flight log file will be a .TXT file not a .DAT file. I just upload to AirData. For uploading here just search for methods.
I have not idea what PCV catch handle looks like or how it is attached.

Mavic

Shared with Dropbox

www.dropbox.com

 

[Tim3459575]

Mavic

 

[Tim3459575]

“Mavic” is a link to the flight record and photos

 

[Yorkshire_Pud]

The log on phantomhelp

DJI Flight Log Viewer | Phantom Help

Upload and view flight logs from your DJI drones

www.phantomhelp.com

 

[Tim3459575]

The log on phantomhelp

DJI Flight Log Viewer | Phantom Help

Upload and view flight logs from your DJI drones

www.phantomhelp.com

thanks Rod

 

[RodPad]

thanks Rod

I don't need the thanks, I just bug people to "GET IT DONE"

Yorkshire_Pud

mavicpilots.com

This guy will most likely give you all your answers.

Now, I just watch.

Rod ..

 

[Yorkshire_Pud]

A quick skim of the log shows 4 warnings about "Vision sensor(s) blocked", at 12, 1075.8, 1126.6 and 1260.8 seconds.
There are two similar warnings, at the start of the flight, but I don't think they are of significance.
I don't know if the "4" are significant but hopefully the likes of @slup @sar104 and @Meta4 will be along to have a look

This guy will most likely give you all your answers.

Thanks for the vote of confidence but sorry I just dabble, it is one or more of the above trio that this person needs, I am too quick to focus on the obvious and a deeper examination by the above reveals something else.

 

[RodPad]

OMG!, Your summing the next upper level guys.

I be watching for a while.

Rod ..

 

[Meta4]

Attached is the flight log. Perhaps this was due to being in C mode ? Hardware issue ? Any advice on what may have caused this and what I should do to prevent it in future, would be much appreciated !

The stalling issues starts to show up at around 14:07 and there are a number of examples after this
Each time the drone is heading around 262 degrees.

When RTH is working to bring the drone back, it is on a heading of 282 degrees.

Obstacle avoidance looks like the most likely explanation, with OA falsely identifying a low, bright sun as an obstacle and OA set to brake.
This can be avoided by turning OA off when not needed, or flying a different course to avoid flying towards the sun.

The blocked VPS messages would have no effect on forwad motion, and your boat's mast and rigging wouldn't have affected anything.

 

[MS Coast]

The stalling issues starts to show up at around 14:07 and there are a number of examples after this
Each time the drone is heading around 262 degrees.

When RTH is working to bring the drone back, it is on a heading of 282 degrees.

Obstacle avoidance looks like the most likely explanation, with OA falsely identifying a low, bright sun as an obstacle and OA set to brake.
This can be avoided by turning OA off when not needed, or flying a different course to avoid flying towards the sun.

The blocked VPS messages would have no effect on forwad motion, and your boat's mast and rigging wouldn't have affected anything.

Do the position and time correspond to a solar azimuth angle of approximately 262 degrees?

 

[Tim3459575]

further background ….

the catch handle partially blocks the forward vision sensors but that hasn’t been a problem in the past.

If you look at the catch handle, (photo previously provided in dropbox link) it’s shaped like a “blade”. this reduces wind resistance when moving forward but increases resistance when moving sideways (it’s a flattened pvc pipe). Before flattening the stick in this manner, I did experience jerky motion, presumably due to air drag on the stick causing the drone to have excessive tilt. Since flattening the stick, I’ve not experienced that issue.

 

[Meta4]

Do the position and time correspond to a solar azimuth angle of approximately 262 degrees?

Yes

 

[HoozierDroneDaddy]

The stalling issues starts to show up at around 14:07 and there are a number of examples after this
Each time the drone is heading around 262 degrees.

When RTH is working to bring the drone back, it is on a heading of 282 degrees.

Obstacle avoidance looks like the most likely explanation, with OA falsely identifying a low, bright sun as an obstacle and OA set to brake.
This can be avoided by turning OA off when not needed, or flying a different course to avoid flying towards the sun.

The blocked VPS messages would have no effect on forwad motion, and your boat's mast and rigging wouldn't have affected anything.

Could the light in the direction he was traveling be from a reflection off the PVC hand catch he has attached. If it is white and the light is reflecting towards the sensor, it would be from below the drone and indicating an obstacle telling it to stop.

 

[Tim3459575]

if it stopped due to a perceived obstacle wouldn’t the flight log indicate that ?