DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

Lost My 3 Pro in the River today...Log Provided

I'm not so sure that CC was started before the uncommanded descent began. Before 261.24 sec full negative throttle had caused the M3 to descend at 5.9 m/sec. At 261.24 secs the negative throttle was changed to neutral and the descent slowed slightly to 5 m/sec. But, it didn't slow any further, indicating the descent was uncommanded. It wasn't until 0.96 secs later that the M3 reported it was in CC.
View attachment 166370
Can you please clarify your conclusions?
1. Was any descent part of the programmed Cruise Control, when Cruise Control was locked and took over?
2. Or, are you suggesting that the subsequent descent was independent of the Cruise Control stick settings that were previously locked?
The first would be pilot error.
The second would suggest a drone error.
 
DJI has responded to my inital fly away case I filed. Now they want a fly away report from my remote. Don't they have all my flight records already that are synced with their servers?
Your synced flight records do not include the detailed data shown in the bound remote's record of the flight, which is necessary for Fly Away coverage, and the synced flight records are also presumably private.
 
  • Like
Reactions: Donzo98
Just paid for the flyaway...$739. At that price DJI Still makes money. Cost of a M3Pro to them is about $90 Thanks everyone for the feedback. You live and learn.
That's the fastest and most economical way to get back in the air, especially if you value your time! Even if you eventually managed to get DJI to cover this as a warranty issue, the many hours, days, and weeks of your time required wouldn’t be worth it, and certainly not if you were unsuccessful. Bird back in the hand. Fly on immediately!
 
Last edited:
Yes it's bound and I didn't WALK AWAY!!! I had controller in one hand and was packing up. Maybe the drone should warn that it's descending or getting close to the ground, PULL UP PULL UP..etc.
Sorry if I mischaracterized the nature of your admitted inattention while the drone drowned itself. "I was not monitoring. I set it down to do something while it returned over the river."

Since you were in CC in Sport Mode, OA was therefore off, by your choice. Hence, why RTH is a preferred autonomous return method. OA stays on.

Speed did contribute to the crash, as it meant OA would be off, and could not prevent flying into the river, no matter the cause of the controlled descent.

Lessons learned.
Fortunate that you had the prescience to fly with the bound remote, so Fly Away coverage would be available!
 
See this video for visual proof that ascension or descension is included in Cruise Control

DJI Mavic 3 / CRUISE CONTROL Tutorial!
At 4:02

However I was already in CC when I hit the stick to drop a little. I did not rehit the CC button to update the change and in the past it has never continued on.
 
Just paid for the flyaway...$739. At that price DJI Still makes money. Cost of a M3Pro to them is about $90 Thanks everyone for the feedback. You live and learn.
They already sent replacement. At the end of the day happy with how DJI handled this of course...
 
  • Like
Reactions: GadgetGuy
Can you please clarify your conclusions?
1. Was any descent part of the programmed Cruise Control, when Cruise Control was locked and took over?
2. Or, are you suggesting that the subsequent descent was independent of the Cruise Control stick settings that were previously locked?
The first would be pilot error.
The second would suggest a drone error.
I was not concluding either 1 or 2. I was just pointing out that an uncommanded descent had started prior to CC starting at 262.2 secs.

Viewing the YouTube you provided I'm inclined to believe that CC had started earlier than 262.2 secs. Starting CC could be a lengthy multistep process and only gets logged after all the steps are complete. Just a theory.

Unfortunately, none of the custom buttons were recorded as having been depressed. That would have given us a more precise CC start time.

I noticed that the descent rate started to slow at about 268 secs. The ultrasonic sensor data was valid and was indicating 12.4 meters. About 2 secs later the ultrasonic sensor data became invalid and the descent rate started to increase.
1690348118139.png
OSD:zSpeed is descent rate since the Z axis points down. I'm guessing the M3 would have stopped descending if it were not for the ultrasonic sensor being confused because the M3 was over water.
 
  • Like
Reactions: GadgetGuy
A little more on Cruise Control from DJI on their forum:

======

To enter the Cruise Control after setting the C1 or C2 button, please do the following:

1. Push the control stick in any direction, and press the Cruise Control button simultaneously. According to the control stick input, the aircraft will fly at the current speed.

2. Before the control stick returns to the center, press the Cruise Control button again, and the aircraft will reset the flight speed based on the current control stick input.

3. Push the control stick after it returns to the center, and the aircraft will fly at the increased speed based on the previous speed. In this case, press the Cruise Control button again, and the aircraft will fly at the increased speed.

Notes:

1. Cruise Control is available in Normal, Cine, and Sport mode, and the function can be enabled in APAS, Free Hyperlapse, and Spotlight (FocusTrack).
2. Cruise Control cannot be started without a control stick input.
3. Cruise Control cannot be started or will exit automatically when near the Max Altitude or Max Distance.
4. Cruise Control cannot be started or will exit automatically when the aircraft disconnects from the remote controller or DJI Fly.
5. Cruise Control cannot be started or will exit automatically after the aircraft senses an obstacle and will hover in place.
6. During RTH or auto landing, the aircraft cannot enter or will automatically exit Cruise Control.
7. Cruise Control will exit automatically when switching between flight modes.
8. The obstacle avoidance in Cruise Control follows the current flight mode. Fly with caution.
9. Cruise Control cannot be started or will exit automatically when the aircraft is approaching to a GEO zone.
10. Cruise Control supports more control stick movements such as ascending the aircraft when flying forward or ascending/descending the aircraft when flying around an object, etc.). If the Cruise Control button is pressed, the current control stick input of the remote controller is locked, and the aircraft can maintain the flight along with the current flight path.
11. When entering cruise control, only the reverse stick movement in the forward direction responds, and the aircraft will move at a lower speed. (For example: If the aircraft flies leftward, moving the stick rightward represents a reverse stick movement. If the aircraft flies forward, moving the stick backward also represents a reverse stick movement.) When other sticks are moved, the aircraft will not respond, and it will not exit cruise control mode.



 
Last edited:
I was not concluding either 1 or 2. I was just pointing out that an uncommanded descent had started prior to CC starting at 262.2 secs.

Viewing the YouTube you provided I'm inclined to believe that CC had started earlier than 262.2 secs. Starting CC could be a lengthy multistep process and only gets logged after all the steps are complete. Just a theory.

Unfortunately, none of the custom buttons were recorded as having been depressed. That would have given us a more precise CC start time.

I noticed that the descent rate started to slow at about 268 secs. The ultrasonic sensor data was valid and was indicating 12.4 meters. About 2 secs later the ultrasonic sensor data became invalid and the descent rate started to increase.
View attachment 166391
OSD:zSpeed is descent rate since the Z axis points down. I'm guessing the M3 would have stopped descending if it were not for the ultrasonic sensor being confused because the M3 was over water.
Thanks for the clarification.

Since the drone was also in Sports Mode while in Cruise Control, which means OA was not available, wouldn't the ultrasonic sensor data be ignored anyway, regardless of any confusion over water, or is the ultrasonic sensor independent of OA, and still active, intended to prevent running into the ground, even in Sports Mode?

CC can also be reset at new current settings, by merely pressing the CC button again. Too bad the CC button depressions are not recorded. Might have shown two depressions.
 
  • Like
Reactions: BudWalker
However I was already in CC when I hit the stick to drop a little. I did not rehit the CC button to update the change and in the past it has never continued on.
There was likely already a descension component on your original CC setting, unbeknownst to you. Since the timing of the CC button depressions are not recorded in your flight log, we'll never know.
 
.....

Since the drone was also in Sports Mode while in Cruise Control, which means OA was not available, wouldn't the ultrasonic sensor data be ignored anyway, regardless of any confusion over water, or is the ultrasonic sensor independent of OA, and still active, intended to prevent running into the ground, even in Sports Mode?

......
At about 270 secs the ultrasonic data becomes invalid (blue background) and the descent rate (green plot OSD:zSpeed) begins to increase.
1690378640919.png

And, at 271 secs the ultrasonic data becomes valid (pink background) and the descent rate (green plot OSD:zSpeed) begins to just noticeably decrease.
1690378934142.png

This behavior strongly suggests that the ultrasonic sensor data is being used to avoid objects below the M3. Either ultrasonic data is not considered to be part of OA or Sport+CC mode has it's own OA rules.
.
 
  • Like
Reactions: GadgetGuy
At about 270 secs the ultrasonic data becomes invalid (blue background) and the descent rate (green plot OSD:zSpeed) begins to increase.
View attachment 166398

And, at 271 secs the ultrasonic data becomes valid (pink background) and the descent rate (green plot OSD:zSpeed) begins to just noticeably decrease.
View attachment 166399

This behavior strongly suggests that the ultrasonic sensor data is being used to avoid objects below the M3. Either ultrasonic data is not considered to be part of OA or Sport+CC mode has it's own OA rules.
.
So, are we to conclude that it couldn't stop the descent fast enough to prevent the drowning in Sports mode, combined with the reflective water confusing the ultrasonic sensors as to the remaining distance to the water surface?
 
  • Like
Reactions: BudWalker
So, are we to conclude that it couldn't stop the descent fast enough to prevent the drowning in Sports mode, combined with the reflective water confusing the ultrasonic sensors as to the remaining distance to the water surface?
That would be my conclusion.
 
  • Like
Reactions: Filmarik
If this is pilot error and a reason to void my DJI Care I'd like to know what I did. As the drone is descending you can clearly see the stick isn't being pushed.

Cruise controlled descent is that even a thing? I"ve flone hundreds of flights in Cruise Control and raised and lowered my height and never did it keep going up or down.
I just threw the question out there, didn't mean it as an accusation. Were you monitoring the flight and did you attempt to take control once it started losing altitude? I suppose I've just been lucky because I've never had any flight problems with any one of the three drones. I have an old Phantom, Inspire 2 (sold it) and Mavic 2. I've accumulated just over 200 hours of flight time. Worked for Mustivista for a few years.
 
Were you monitoring the flight and did you attempt to take control once it started losing altitude?
The incident could have been avoided if the screen had been checked once the drone was in cruise control.
 
At the end of the day I don't expect DJI to do ****. All the years I've been flying with their warranties and everything I see here they wiggle out every time. Several years ago I had a stick issue on a controller that sent my drone into the river. The stick was showing it was stuck in the down position even though the stick was physically fine on the controller. Not only would DJI not cover the drone after confirming the controller malfunctioned, they charged me to fix the controller.
Everybody has different experiences with them, I've used the warranty and refresh once and had to py once when I didn't renew the refresh and crashed afterward. Each time they were good. Maybe it was because the mini 2 is so cheap but I have seen hundreds of positive reviews on this site as well as a few negative ones. I think they are doing a good job...
 
When I was working for Multivista they had an Inspire 2 crash and thought it was a total. Sent it in and they repaired it for $1200.00, a lot cheaper than a new Inspire 2.
 
They already sent replacement. At the end of the day happy with how DJI handled this of course...
Replacement Drone and battery arrived today. Now I need to decide whether to sell the M3P or hang onto it. Quite honestly I'm not finding anything it does better then the Air 3 for my needs right now. I like the Air 3 a lot.
 
  • Like
Reactions: Roberto Smera
Short flight to photograph a ship, was on my way back. Set cruise control and was over 100 feet..then it descends, you can see the controller sticks not being touched. Standing there waiting for the return and hear the controller say ATTI MODE and I look at screen and the drone is dropping down fast right into the water. I reported this here a few months ago I had the bird start to drop altitude without touching the sticks but this drone hasn't done it until today. I've had tons of care warranties, never have used one for replacement or flyaway. How do I go about getting DJI to deal with this since I don't have the drone to send them? Here is my flight log. Any thoughts and suggestions would be appreciated.

Same thing happened to me over Lemon Bay, FL. All of a sudden, without any input, she landed in the water. It appeared to be a controlled landing and there was nothing I could do about it. I was about 40 feet from the drone.
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
131,141
Messages
1,560,327
Members
160,113
Latest member
seacorope