Mavic Pro2 Falls from Sky

[dewster]

I’m sure there will be other incidents if it turns out to be hardware related.

 

[Former Member]

So, the jury is still out on this one. No analytical proof, or at least incomplete proof, of what mechanical failure caused the drone to go down. I trust that the OP is not making up a story that it was a mechanical failure.

Im guessing that DJI will cover it, but the new OP might have to be a squeaky wheel.

 

[sar104]

So, the jury is still out on this one. No analytical proof, or at least incomplete proof, of what mechanical failure caused the drone to go down. I trust that the OP is not making up a story that it was a mechanical failure.

Im guessing that DJI will cover it, but the new OP might have to be a squeaky wheel.

Having seen the DAT and txt logs I don't doubt the OP's account. And it was almost certainly not a mechanical failure - it looks like a processor problem to me.

 

[Former Member]

Having seen the DAT and txt logs I don't doubt the OP's account. And it was almost certainly not a mechanical failure - it looks like a processor problem to me.

Would not a processor problem be considered a mechanical problem?

 

[sar104]

Would not a processor problem be considered a mechanical problem?

Probably not since the 1960s, when mechanical computers were replaced by electronic processors.

 

[Former Member]

Probably not since the 1960s, when mechanical computers were replaced by electronic processors.

Thank you for the explanation. In a typical drone, I guess there are many systems that operate it and can fail. Electronic, mechanical, software, etc. In this case we are talking about a processor failure. Im not a scientist, but would it be fair to say that a processor problem led to a mechanical failure in this case? Or the processor failed, and shut down the functioning mechanical parts.

 

[sar104]

Thank you for the explanation. In a typical drone, I guess there are many systems that operate it and can fail. Electronic, mechanical, software, etc. In this case we are talking about a processor failure. Im not a scientist, but would it be fair to say that a processor problem led to a mechanical failure in this case? Or the processor failed, and shut down the functioning mechanical parts.

I really guessing here, but the anomalous log entries and abrupt end suggest to me that FC processes were failing. If it eventually shut down then everything stops, which is not a mechanical failure per se - i.e. the mechanical systems are still good. Until it hits the ground of course. By mechanical systems I really just mean the airframe, motors, props etc..

 

[Former Member]

I really guessing here, but the anomalous log entries and abrupt end suggest to me that FC processes were failing. If it eventually shut down then everything stops, which is not a mechanical failure per se - i.e. the mechanical systems are still good. Until it hits the ground of course. By mechanical systems I really just mean the airframe, motors, props etc..

Do you think that's why they have made redundant IMU's in the Mavic line, to reduce the possibility of a catastrophe if one were to fail? I have flown a few Mavics that have had 1 IMU fail and the drone flew ok, but the drone lost the redundancy of a second IMU.

 

[sar104]

Do you think that's why they have made redundant IMU's in the Mavic line, to reduce the possibility of a catastrophe if one were to fail? I have flown a few Mavics that have had 1 IMU malfunction, but the drone lost the redundancy of a second IMU.

I think they are still trying to figure out the optimum configuration. The P4 has twin IMUs and twin compasses, the MP has one IMU and twin compasses, the MA has twin IMUs and one compass (plus a vision compass that remains somewhat mysterious) and the M2 has the same as the MA. I suspect that IMU failure is more common than compass failure (it's generally interference that messes up the compass) and with twin compasses it seemed that the FC never knew which to trust anyway.

 

[Cyberpower678]

Having seen the DAT and txt logs I don't doubt the OP's account. And it was almost certainly not a mechanical failure - it looks like a processor problem to me.

More like the IMU froze and stopped recording forcing a switch to other IMU which also stopped for a moment.

 

[sar104]

More like the IMU froze and stopped recording forcing a switch to other IMU which also stopped for a moment.

That would certainly bring it down, although something also stopped the FC from communicating with the remote.

 

[Cyberpower678]

That would certainly bring it down, although something also stopped the FC from communicating with the remote.

I mentioned this at a theory in a different thread, which is shaping to be true. When the something goes awry internally that can't be resolved, it just shutdown mid-flight. By it, I mean the FC. That's why the RC disconnected.

 

[sar104]

I mentioned this at a theory in a different thread, which is shaping to be true. When the something goes awry internally that can't be resolved, it just shutdown mid-flight. By it, I mean the FC. That's why the RC disconnected.

Maybe so. It's unfortunate that we can't read the aircraft DAT files - they might shed more light on this.

 

[sar104]

I think they are still trying to figure out the optimum configuration. The P4 has twin IMUs and twin compasses, the MP has one IMU and twin compasses, the MA has twin IMUs and one compass (plus a vision compass that remains somewhat mysterious) and the M2 has the same as the MA. I suspect that IMU failure is more common than compass failure (it's generally interference that messes up the compass) and with twin compasses it seemed that the FC never knew which to trust anyway.

I need to clarify this statement. @Thunderdrones pointed out to me that the GO app sensors screen on the Mavic Pro shows two rate gyro and accelerometer packages such as are seen other models with two IMUs. However, the DAT files only indicate one set of IMU calculations (IMU_ATTI(0):xxx) suggesting just one IMU. So in the case of the Mavic Pro it appears that there are twin sensor packages for redundancy, but only one actual IMU performing calculations, and it chooses which sensor data to use.

This is in contrast to the Mavic Air and Mavic 2 that record two sets of IMU output, IMU_ATTI(0):xxx and IMU_ATTI(I):xxx.

 

[IdahoJP]

Thank you all for the input, The drone went off to DJI today will keep you informed of the progress. I put in with the drone some of the comments from this forum. Will see if that helps or hurts me in the long run.

 

[duelingglove]

Aw I’m about to get back into flying and wonder if this is an isolated incident. My lil original Mavic crashed once but after I repaired it hasn’t had any other issues

 

[Dextra]

I've assessed ten cases this week of battery overheating and batteries popping off mid-flight due to extreme overheating, ESC failure's are another common problem seen thus far with both models. ESC's reaching max temp, then shutting down. Most of the drones have fallen into the sea, lake, river, others like this above, simply crashing to the road/path - done - and - dusted. I don't need to take a look at the log. Your description was enough to suggest its yet another battery and/or ESC failure. Raise a claim with DJI. They'll send another one out. I was going to buy one but have waited for numerous faults to arise which they have and likely more will follow too.

 

[Cymruflyer]

I ordered the Mavic 2 Pro and now I'm a little worried. My trusty old Mavic Pro worked flawlessly for almost a year and a half until I sold it on last week. Now I'm wondering if I should have waited to get one in about 3 months. I suppose the amount out there compared to the ones that have had problems causing them to drop out of the air are at a percentage point that I should not worry too much.

I was wondering if there might be a correlation to the manufacturer's serial numbers on these aircraft, to those that are having problems? Like all the ones having problems are from a particular serial batch number maybe??? Has anyone thought to collect the serial numbers from those that have had or are having problems, to see if anything can be garnered from that data too, as simple as that might be to check? Maybe a particular run are now having a history of failures.

I can also say that I take my hat off to all you amazing gurus here who are able to understand and work out so much info about these incidents, you guys are like drone info Gods to me. And although I hope I never need your assistance, I want to say a huge thank you to everyone who offers their time and expertise to the likes of us simple drone flyers in this community.

 

[khooper802]

My First time posting on this forum but I need to know what Happened. I did a small fight from my house had the drone in sight when two warnings popped up and the drone dropped like a rock onto the road. I know pilot error is most of crashes but I did nothing wrong that I can see. You tell Me. The link is enclosed.

DJI Flight Log Viewer - PhantomHelp.com

After looking at other post is over heating the cause?

First and only mistake I think you made was buying the Mavic 2 when it came out. It seems like every drone that is released seems to have an issue. Just like the Mavic air when it came out. All I saw was that the bird was falling out of the sky every other flight, same with some phantoms. I waited about 4 months to get my Mavic Pro I and I have had only one issue with it but was an error on my part for not letting the IMU calibrating correct but other than that no issues. DJI also announced a software update that I think was taking place soon so I'm sure there they address most issues that you "Testers" have been reported and sending to them. Hopefully they fix your drone as I see you did nothing wrong and were over 200 feet high so its not like you ran into anything.

 

[WParm]

This is a reason that I am typically not an early adopter of any type of technology. Waiting until the glitches are worked out saves me a ton of headaches.