Does the Mavic Have a Karma Problem? (Inflight Shutdowns)

[JLane]

It seems some users are starting to worry about the possibility of their Mavic dropping from the sky, given the recent incidents of unexplained inflight shut downs. Some say it's nothing to worry about, whiles others are starting to feel jittery.

I decided to post this thread to share with you all the data that I've been gathering on this issue. Shortly after I started to see a few of these cases being reported on MavicPilots and the DJI forums, I decided to start logging as much information as I could, fearing there could be a trend. My data so far includes 14 reported incidents. One of those happened back in February, while all the other 13 were posted in April. Now, I'll be the first to admit that 14 reported cases is not a huge number, given the thousands of Mavics that must be in the air every day. However, 13 reported cases within just a few weeks does not seem insignificant, and we don't know how many others have never been reported.

In pretty much all of these cases, there has been no real cause attributed to the crash, aside from the obvious complete power cut. @BudWalker and @Robbyg have both looked at several of the available .dat files, and while there was an apparent anomaly seen in some of them, it isn't present (or missing, to be more accurate) in all cases. DJI seems to be covering these recent crashes under warranty, indicating there was no pilot error.

I don't want to be alarmist, and that's part of why I didn't post my findings earlier, but I now think perhaps it would be good for everybody to be able to track this data as well. I will leave the data here for others to decide what they think.

In case the iFrame embedding doesn't work, here's a link to the data: Mavic Karma - Airtable

 

[Former Member]

Good statistics. Just trying to draw a conclusion from it.

14 reports so far.
4 had no build date, so not sure if March production was involved
4 had dats posted
2 of the 4 dats posted had dat anomalies
1 of the 2 that had dat anomalies crashed after updating firmware

Of the 14 incidents, 4 dats could be retrieved and interpreted
Of the 4 dats that could be interpreted, 2 had dat anomalies.

Does that sound right?

Do we consider flyaways "falling out of the sky?" Just trying to clarify what we mean when someone says their Mavic fell from the sky.

 

[JLane]

Good statistics. Just trying to draw a conclusion from it.

14 reports so far.
4 had no build date, so not sure if March production was involved
4 had dats posted
2 of the 4 dats had dat anomalies
1 of the 2 that had dat anomalies crashed after updating firmware

Of the 14 incidents, 4 dats could be retrieved and interpreted
Of the 4 dats that could be interpreted, 2 had dat anomalies.

Does that sound right?

That sounds about right. Except I wouldn't say only 4 dats COULD be retrieved. It's more like 4 data WERE retrieved, and in other cases the poster just didn't try, or hasn't posted it.

Also, I can't say I'm 100% up-to-date on all the data. It's sometimes hard to follow it all closely.

 

[Robbyg]

The Mavics that were analyzed all had the power to their motors disconnected in mid flight. It's either bad battery latching by the user, Defective latches by the manufacture or one of the circuit boards is somehow faulty. Some have speculated that it is firmware related. I doubt that because if it was the number of sudden power loses would be much higher. This seems to be an issue with a limited batch of Mavics.

Rob

 

[Former Member]

That sounds about right. Except I wouldn't say only 4 dats COULD be retrieved. It's more like 4 data WERE retrieved, and in other cases the poster just didn't try, or hasn't posted it.

Also, I can't say I'm 100% up-to-date on all the data. It's sometimes hard to follow it all closely.

Thanks for your hard work on this.

 

[Former Member]

The Mavics that were analyzed all had the power to their motors disconnected in mid flight. It's either bad battery latching by the user, Defective latches by the manufacture or one of the circuit boards is somehow faulty. Some have speculated that it is firmware related. I doubt that because if it was the number of sudden power loses would be much higher. This seems to be an issue with a limited batch of Mavics.

Rob

Gets me to thinking... "power to their motors disconnected in flight." Maybe battery overheated and failed, causing drop in voltage and not enough power to autoland. Any data on battery temps?

Also, main board overheating could cause ESC thermal failure, and that would either supply uneven voltage to one or more of the motors, or just stop supplying voltage altogether resulting in an uncontrollable descent. I race RC cars, and sometimes if its really hot outside, my ESC would thermal and shut down all power to the motor until it cooled down.

Improper or faulty battery latching could be a culprit, I always make sure to hear a click and then try to pull the battery out once it's latched in.

 

[JLane]

Gets me to thinking... "power to their motors disconnected in flight." Maybe battery overheated and failed, causing drop in voltage and not enough power to autoland. Any data on battery temps?

Also, main board overheating could cause ESC thermal failure, and that would either supply uneven voltage to one or more of the motors, or just stop supplying voltage altogether resulting in an uncontrollable descent. I race RC cars, and sometimes if its really hot outside, my ESC would thermal and shut down all power to the motor until it cooled down.

Improper or faulty battery latching could be a culprit, I always make sure to hear a click and then try to pull the battery out once it's latched in.

I'm not as knowledgable as you are on this, but I would assume that an ESC shutdown would probably not cause a full shutdown of the whole system, would it? One common thread in all these cases is that the logged data just stops abruptly. Users report simultaneous RC disconnection and motors cutting, while the logs show nothing other than an abrupt end. Although in some cases I believe BudWalker has shown a very sudden drop in voltage was logged just before the logs stopped.

 

[Former Member]

I'm not as knowledgable as you are on this, but I would assume that an ESC shutdown would probably not cause a full shutdown of the whole system, would it? One common thread in all these cases is that the logged data just stops abruptly. Users report simultaneous RC disconnection and motors cutting, while the logs show nothing other than an abrupt end. Although in some cases I believe BudWalker has shown a very sudden drop in voltage was logged just before the logs stopped.

A sudden drop in voltage usually occurs when lipos are near depletion. They start out peppy, then during flight they linearly drop until around 20-25% then they can dump on you, meaning drop drastically that last 20%. I always try to plan on landing with at least 20% battery left because I know lipos can provide uneven power when they are down that low and possibly dump altogether which means one of the systems is going to stop functioning, in this case the esc and motors because it draws the most current.

I think the primary focus of the power system is getting power to the ESC and motors to maintain flight. A power brownout could cut power to the cam and other systems before it cuts power to the motors and esc. If you lose video but it keeps flying, it could be a variety of things including a brownout. But if it happens all at the same time, I would think it means total battery failure, or main board failure which would include ESC and gimbal. The battery is of no use without a functional main board, and vice versa. It would be nice to be able to find some of those drones that are MIA and do a battery test on them or retrieve dat files.

I try to warn people not to fly when it gets too hot outside because your batteries are not meant to operate when they exceed 120° and can fail and puff if over 140. The battery is in a very tight compartment, and doesnt get good airflow at the bottom, beneath which is the main board which gets very hot on it's own.

Im trying not to point fingers as if all these "falling from the sky" incidents are operator error, we all know drones flyaway, and sometimes crash for unknown reasons. But having a little more understanding of how to treat your drone and battery could go a long way in helping prevent these types of unexplained phenomena.

 

[JLane]

A sudden drop in voltage usually occurs when lipos are near depletion. They start out peppy, then during flight they linearly drop until around 20-25% then they can dump on you, meaning drop drastically that last 20%. I always try to plan on landing with at least 20% battery left because I know lipos can provide uneven power when they are down that low and possibly dump altogether which means one of the systems is going to stop functioning, in this case the esc and motors because it draws the most current.

I think the primary focus of the power system is getting power to the ESC and motors to maintain flight. A power brownout could cut power to the cam and other systems before it cuts power to the motors and esc. If you lose video but it keeps flying, it could be a variety of things including a brownout. But if it happens all at the same time, I would think it means total battery failure, or main board failure which would include ESC and gimbal. The battery is of no use without a functional main board, and vice versa. It would be nice to be able to find some of those drones that are MIA and do a battery test on them or retrieve dat files.

I try to warn people not to fly when it gets too hot outside because your batteries are not meant to operate when they exceed 120° and can fail and puff if over 140. The battery is in a very tight compartment, and doesnt get good airflow at the bottom, beneath which is the main board which gets very hot on it's own.

Im trying not to point fingers as if all these "falling from the sky" incidents are operator error, we all know drones flyaway, and sometimes crash for unknown reasons. But having a little more understanding of how to treat your drone and battery could go a long way in helping prevent these types of unexplained phenomena.

If you want to get stuck in and get your hands dirty, I'd recommend this recent case as a good one to take a look at: Mavic's engines stopped for no reason and fall down on the floor

The Mavic appears undamaged (low height crash) and the owner continues to fly it. The aircraft and battery are still fully intact, so whatever data you want should be available. The owner seems very willing to help answer any questions, as he's very interested in finding out what the root cause was.

ps. All or most of these cases happened with plenty of charge left in the battery. Perhaps I should try to add that as a data point.

 

[Former Member]

If you want to get stuck in and get your hands dirty, I'd recommend this recent case as a good one to take a look at: Mavic's engines stopped for no reason and fall down on the floor

The Mavic appears undamaged (low height crash) and the owner continues to fly it. The aircraft and battery are still fully intact, so whatever data you want should be available. The owner seems very willing to help answer any questions, as he's very interested in finding out what the root cause was.

ps. All or most of these cases happened with plenty of charge left in the battery. Perhaps I should try to add that as a data point.

This was his first flight ever with the drone, although he had taken off and landed twice before on this flight.

DJI's conclusion was "battery cut off."

He makes a comment that, "Funny thing is that it will actually still turn on and make the happy DJI sound." That means the main board and battery were ok after the crash and were probably ok before the crash.

In any event, he won his claim with DJI. If he had 20 flights on it and this happened, I would take a deeper look into it. But a Mavic landing or crashing itself on its first flight leaves too much to the imagination of what really happened.

Is there any way to tell what his critical battery level was set at from the logs?

 

[JLane]

This was his first flight ever with the drone, although he had taken off and landed twice before on this flight.

DJI's conclusion was "battery cut off."

He makes a comment that, "Funny thing is that it will actually still turn on and make the happy DJI sound." That means the main board and battery were ok after the crash and were probably ok before the crash.

In any event, he won his claim with DJI. If he had 20 flights on it and this happened, I would take a deeper look into it. But a Mavic landing or crashing itself on its first flight leaves too much to the imagination of what really happened.

Is there any way to tell what his critical battery level was set at from the logs?

I think you're talking about the thread that was linked to in the first post of the thread that I gave you. The one I'm referring to (by user odradek) never mentioned it being his first flight, he had 60% battery left at the time of the crash, and he has yet to hear back from DJI about his claim. In fact, I'm not even sure if he's making a warranty claim, as his Mavic still works fine.

 

[AyeYo]

Can it be verified that these aren't related to a battery firmware mismatch?

 

[Former Member]

I think odradek's battery wasnt seated well and probably popped out and caused power failure. At 40 seconds into the video, there is a knocking sound, possibly controller sticks but no aircraft movement during the knocking sounds. Then he climbed pumping the throttle, and then at 43 seconds there were more knocking sounds, then at 50 one last knocking sound and then the crash.

Could be the props hitting the battery which had come loose, then just came out. He didnt remember if the drone was on or off when he picked it up, so no smoking gun there. If it was off, Im thinking battery came loose. If it was still on, maybe he hit something like a tree branch. Wonder what that knocking noise was?

 

[Majingear]

Hey guys,

I came here from an other forum.

I got the same problem as here mentioned. My Mavic just crashed mid air, over a big field. No obstacles around.

Here you can find the thread: Mavic Crash aus 10m - was ist die Ursache/wie gehe ich jetzt vor? - DJI Mavic

Its in German, but Google translator should work pretty good.

If you are interested you can check out the FLY012 Log from this dropbox link: Dropbox - Maviic

 

[Former Member]

Hey guys,

I came here from an other forum.

I got the same problem as here mentioned. My Mavic just crashed mid air, over a big field. No obstacles around.

Here you can find the thread: Mavic Crash aus 10m - was ist die Ursache/wie gehe ich jetzt vor? - DJI Mavic

Its in German, but Google translator should work pretty good.

If you are interested you can check out the FLY012 Log from this dropbox link: Dropbox - Maviic

I have read your post from the other forum and it is possible that your battery was not properly mounted since it was found a few meters away from where the mavic crashed.

 

[Majingear]

I have read your post from the other forum and it is possible that your battery was not properly mounted since it was found a few meters away from where the mavic crashed.

thats true, thats the 2nd possibility what caused it to crash. I just wanted to share more info to this possible manufacturing problem

 

[Former Member]

thats true, thats the 2nd possibility what caused it to crash. I just wanted to share more info to this possible manufacturing problem

Can you tell me what the temperature was outside when you were flying? You said the Mavic was hot when you picked it up, but the battery wasnt so hot. Mavics do run hot at the heatsink, so hot is normal. If it was too hot to touch, then it could have been a problem.

 

[Former Member]

Hey guys,

I came here from an other forum.

I got the same problem as here mentioned. My Mavic just crashed mid air, over a big field. No obstacles around.

Here you can find the thread: Mavic Crash aus 10m - was ist die Ursache/wie gehe ich jetzt vor? - DJI Mavic

Its in German, but Google translator should work pretty good.

If you are interested you can check out the FLY012 Log from this dropbox link: Dropbox - Maviic

I have read your post from the other forum and it is possible that your battery was not properly mounted since it was found a few meters away from where the mavic crashed.

 

[Majingear]

Can you tell me what the temperature was outside when you were flying? You said the Mavic was hot when you picked it up, but the battery wasnt so hot. Mavics do run hot at the heatsink, so hot is normal. If it was too hot to touch, then it could have been a problem.

its was pretty cold outside, like 10-13°C

Edit: + cold wind

 

[Former Member]

its was pretty cold outside, like 10-13°C

Im thinking it was probably not an overheating issue with either the battery or main boards, unless your fan wasnt working.

I hope DJI is fair with you and you get a replacement Mavic at no charge.