Hello,
I flew my mavic today for the first time, I flew a phantom for several months, but at some point flew away without being able to control it. I still can not understand what happened.
If you read the data here http://www.phantomhelp.com/LogViewer/LOMNIIU8G0FCG7EZ2X3R/, can you tell me if there is a pilot error or is there a fault at the unit?
It's a flaw of the latest update? I see there an error about no fly zone, but it not a no-fly zone.
After a time when I thought I lost forever, got into gps mode and I managed to activate RTH.
I am extremely confused, phantom flies there without any error.
Thanks for any information or recommendation.

Was your takeoff location on concrete by any chance? It may have been due to a magnetic distortion.

The best way for us to determine what happened is to examine the DAT file from the flight. Please use these instructions to access the file, then upload it to DropBox and provide a link to it in this thread and we may be able to tell you what happened.

You seemed to be hovering for quite awhile, and when you went to fly away your aircraft just took off , you seem to have serious magnetic interference, you can't tell from the logs where you picked this up whether it was on the ground as droneflying pointed to , you had compass error which then seem to cause your aircraft to take off probably in a different direction than you wanted it to.
Miraculously it seemed to correct itself and go back to gps and fly home.

Droneflying is right to confirm what exactly happened you really need to upload dat file, he knows a good guy who can read dat bud walker he should be able to tell you what happened.

I would also sync your flight log to dji cloud and try get them to see if you need to send it in for repair.
And no it wasn't pilot error.

Yeah, I took off from a concrete pavement but it does not have metal inside. Can this be a problem?
Pavement looks like this
Here you can find the latest .dat file
https://www.dropbox.com/s/cve4lru4yoz79ii/FLY016.DAT?dl=0
Thank you.

danbale Posted at 2017-6-9 13:52
Yeah, I took off from a concrete pavement but it does not have metal inside. Can this be a problem?
Pavement looks like this
Here you can find the latest .dat file

That doesn't appear to be the correct DAT file, so please check again for the one that matches the time of the flight where you had the problem. It will be quite a bit larger than the one that you uploaded, and once you do I'll look at and see what I can find. I may be able to diagnose the problem myself, but if I'm not able to then as Hallmark mentioned BudWalker will definitely be able to figure out what happened.

danbale Posted at 2017-6-9 13:52
Yeah, I took off from a concrete pavement but it does not have metal inside. Can this be a problem?
Pavement looks like this
Here you can find the latest .dat file

Your right no steel in those pierced blocks. Where you flying close to or over something that may have caused magnetic interference?

DroneFlying Posted at 2017-6-9 14:31
That doesn't appear to be the correct DAT file, so please check again for the one that matches the time of the flight where you had the problem. It will be quite a bit larger than the one that you uploaded, and once you do I'll look at and see what I can find. I may be able to diagnose the problem myself, but if I'm not able to then as Hallmark mentioned BudWalker will definitely be able to figure out what happened.

Okay, this is the previous file, it's bigger.
https://www.dropbox.com/s/l6us6vpyd0m009t/FLY015.DAT?dl=0

hallmark007 Posted at 2017-6-9 14:37
Your right no steel in those pierced blocks. Where you flying close to or over something that may have caused magnetic interference?

Well, I did not fly, just hovering over myself, when it suddenly started to slowly turn around in a small circle, then went away.

danbale Posted at 2017-6-9 14:56
Well, I did not fly, just hovering over myself, when it suddenly started to slowly turn around in a small circle, then went away.

So it looks like its compass problem , just need to find out what caused it.

What have I done:
- I charge the battery of the controller, and the unit
- I did all the updates
- I went out to a small hovering test, Was the first flight, I did not want to fly more than 30m up.


What I did not do:
- I did not calibrate the compass
- I did not calibrate IMU

I assumed that it was needed, dji go would have sent a message
Should I do this after the update?

danbale Posted at 2017-6-9 15:06
What have I done:
- I charge the battery of the controller, and the unit
- I did all the updates

You did everything right , there is one way of checking your compass before flight. Bottom left hand corner small red triangle if this is pointing in a different direction than the heading of your aircraft then you need to move your aircraft to another location.
But you can still get magnetic interference when your in the air,

Anytime you have a new machine or move the machine to a different location that your not sure about you should calibrate the compass - possibly the last place it was calibrated was overseas.  

BoomStrike Posted at 2017-6-9 18:25
Anytime you have a new machine or move the machine to a different location that your not sure about you should calibrate the compass - possibly the last place it was calibrated was overseas.

I thought you are only supposed to calibrate the compass when you are asked to do so.

Please don't encourage people to calibrate when they don't need to.  Unnecessary calibration can lead to a more serious problem in some cases.

BoomStrike Posted at 2017-6-9 18:25
Anytime you have a new machine or move the machine to a different location that your not sure about you should calibrate the compass - possibly the last place it was calibrated was overseas.

Do not calibrate unless you are told to do so by Go app.

Honestly you were lucky to get the drone back. You started with a battery only half charged, you had numerous yaw errors, compass errors, you were warned there were obstacles on the RTH, you lost connection for a time and it eventually landed with 15% battery.
There is nothing in the log about no fly zones.

BoomStrike Posted at 2017-6-9 18:25
Anytime you have a new machine or move the machine to a different location that your not sure about you should calibrate the compass - possibly the last place it was calibrated was overseas.

That's a total myth it clearly tells you in your manual NOT to calibrate compass unless prompted. You can travel 1000 miles and if your compass is good then leave it alone it's good.

Calibration of a compass is a very delicate process and it can easily add a discrepancy more easily than to fix things. Calibration should be done only if there is some change in the hardware, some sensor degradation or displacement, some permanent magnetics on the mavic parts etc. Otherwise it is a bad thing to do. If you have a warning of a magnetic interfeerence try to find a place without it. It is absoluetly awfull idea to calibrate near such a place.

Jenee 2 Posted at 2017-6-9 22:10
Honestly you were lucky to get the drone back. You started with a battery only half charged, you had numerous yaw errors, compass errors, you were warned there were obstacles on the RTH, you lost connection for a time and it eventually landed with 15% battery.
There is nothing in the log about no fly zones.

Hello, I flew very little today to two different places in the open field, every time I stood at 10-12 m and waited to see what was happening, in each case I received the message "Current RTH route will pass a No -Fly zone. Pay attention to the aircraft's position to avoid RTH fail", before the unit went into ATTI mode, then moved easily to one side. Of course, those areas are not restricted as NO-Fly zones.

danbale Posted at 2017-6-10 07:02
Hello, I flew very little today to two different places in the open field, every time I stood at 10-12 m and waited to see what was happening, in each case I received the message "Current RTH route will pass a No -Fly zone. Pay attention to the aircraft's position to avoid RTH fail", before the unit went into ATTI mode, then moved easily to one side. Of course, those areas are not restricted as NO-Fly zones.

I would be more worried about mag interference than NFZ , did you check your compass arrow on your map small red triangle, to see if it had the same heading as your Aircraft.

I know droneflying is checking out your dat file so he should have some info for you soon.

danbale Posted at 2017-6-10 07:02
Hello, I flew very little today to two different places in the open field, every time I stood at 10-12 m and waited to see what was happening, in each case I received the message "Current RTH route will pass a No -Fly zone. Pay attention to the aircraft's position to avoid RTH fail", before the unit went into ATTI mode, then moved easily to one side. Of course, those areas are not restricted as NO-Fly zones.

I've looked at the second DAT file you posted and to be honest I haven't yet been able to figure out what's happening. I don't believe it's due to a problem with your takeoff location, and the fact that you've experienced the same problem flying in other locations strongly suggests that something else is the cause. I'll continue to look at it and try to figure out what's going on, and I've been in contact with BudWalker who's also going to look into it when he gets a chance.

danbale Posted at 2017-6-10 07:02
Hello, I flew very little today to two different places in the open field, every time I stood at 10-12 m and waited to see what was happening, in each case I received the message "Current RTH route will pass a No -Fly zone. Pay attention to the aircraft's position to avoid RTH fail", before the unit went into ATTI mode, then moved easily to one side. Of course, those areas are not restricted as NO-Fly zones.

There are two things I would do.
Calibrate the compass and go into the DJI GO 4 settings and update the no fly zones data file in case there is a problem with yours.

I looked at this flight. Actually, it's the second flight in the .DAT. There isn't any evidence to suggest that the launch site was geomagnetically distorted.
This flight is very odd, I can't provide any compelling answers. There seem to be 3 possibilities. 1) The compass isn't calibrated properly. 2) There is some external device attached to the Mavic. 3) It's broke - it's gotta make a trip to DJI repair.

At 623 secs the pitch and roll started oscillating and finally got bad enough that at 683 secs the FC switched to ATTI. Apparently the switch to ATTI arrested the oscilllations. I've never seen anything like this. There are no control  inputs during this interval with the exception of some throttle input at the very end. This spontaneous oscillation is almost like the gain on one of the control loops isn't correct; but that's extremely unlikely since there would be other Mavic incidents with the same behavior. See the first attached

But that isn't the only unusual aspect of this flight. It would appear that the compass is either not calibrated properly, or, there has been something added to the Mavic that is distorting the geomagnetic field. Between 465 and 550 the Mavic is rotated CCW. See 2nd attached. Note appearance of magYaw and Yaw and the way magYaw leads Yaw and then follows Yaw. This looks exactly like the data in some of the experiments I did with my P3 where a bad calibration was obtained on purpose. This could also be caused by the addition of an external device that contains some ferrous material. Has an external device been attached?

BudWalker Posted at 2017-6-11 05:06
I looked at this flight. Actually, it's the second flight in the .DAT. There isn't any evidence to suggest that the launch site was geomagnetically distorted.
This flight is very odd, I can't provide any compelling answers. There seem to be 3 possibilities. 1) The compass isn't calibrated properly. 2) There is some external device attached to the Mavic. 3) It's broke - it's gotta make a trip to DJI repair.

Is there a possibility that the EMI created a false control input to the flight controller from within the AC?
Can this explain the no remote controller inputs from showing up in the Flight data?
   

BudWalker Posted at 2017-6-11 05:06
I looked at this flight. Actually, it's the second flight in the .DAT. There isn't any evidence to suggest that the launch site was geomagnetically distorted.
This flight is very odd, I can't provide any compelling answers. There seem to be 3 possibilities. 1) The compass isn't calibrated properly. 2) There is some external device attached to the Mavic. 3) It's broke - it's gotta make a trip to DJI repair.

Thanks for your time and the analysis presented.
I apologize not write very well in English, I'm helped by google translate.
I'm not familiar with the technical details of flight modes, but I can tell you that mavic is brand new, it was the first real flight.
What I did was raise the mavic to several tens of meters, stay in the air for a few minutes, make a 360 degree circular motion, then I stopped. Mavic continued the circular motion without command.
Mavic interrupted the connection with the remote. The moment he turned to the left and went to my right, I could not control the direction. I could control the movement of the camera, and perhaps the height of the flight, I can not be sure about it because it came out of my sight, but not the direction of the flight.
After about 3 minutes, he managed to get along with the satellites, but he was out of the distance limit set in the dji go software and started the RTH procedure.
What I have tried to do is to increase the speed of return, running against Mavic and landing manually when it came over me.
And to answer your questions, I did not attach any devices to MAVIC.
It was a simple test flight.
I have already written in the previous messages, I did not calibrate the compass. I do not know if I have to do that. Would you recommend me to do this now? Under what conditions?
What is the probability of the unit being defective? Being in possession of me for several hours until this flight, I'm thinking of sending it back to the store to get another one.
Is this a fault that can be remedied me safe? Would you recommend sending it back to the store?

danbale Posted at 2017-6-11 10:15
Thanks for your time and the analysis presented.
I apologize not write very well in English, I'm helped by google translate.
I'm not familiar with the technical details of flight modes, but I can tell you that mavic is brand new, it was the first real flight.

I also use Google translate when I'm conversing with my German colleagues. It's OK, but limited. I'm sure your English is much better than any attempt that I would make to converse in your language.

Your description of events is not supported by the data in a few aspects. But, it doesn't matter because the problem is not related to the events you described.
1) there was no disconnect between the Mavic and the RC,
2) The 310 degree CCW turn starting at 466 secs was the result of negative rudder input. When the rudder input stopped the turn stopped. The turn didn't continue after this.
3) The Mavic was always controllable. When the Mavic switched from GPS+ATTI to ATTI at 681 secs the Mavic began to drift with the wind. While in ATTI very little control input was provided but the Mavic was moving anyway because of the wind. Maybe you were expecting the Mavic to hover in place and the drifting gave the impression that it was uncontrollable?

Anyway, as I said, these points are not relevant. This flight had other more serious problems that I described in my first post. I.e. the diverging oscillations starting at 623 secs.

I'm reluctant to provide any advice because I don't know the cause of the incident. You could calibrate the compass but that may only provide a false sense of security. I.e. it may fly OK for several flights but then fail because it was going to do that anyway. You mentioned the possibility of exchanging the Mavic. Maybe this is your best option.

Good Luck.

bombSaWay Posted at 2017-6-11 07:40
Is there a possibility that the EMI created a false control input to the flight controller from within the AC?
Can this explain the no remote controller inputs from showing up in the Flight data?

I'm not sure I understand the question. What EMI are you referring to?

In reference to the comment about no controller inputs I suppose you may be using CsvView/DatCon to look at the .DAT. The Mavic's new firmware (01.03.0800) has changed that part the .DAT record structure containing the controller inputs. There will be a new version of CsvView/DatCon in a few days that works with the new structure. PM me if you want a beta version before then.

Alternatively, you could use .txt file that can be downloaded from the PhantomHelp link provided by the OP. That .txt can then be viewed using CsvView

Hello,

I come with an update to my situation.

At the recommendation of the DJI representatives, I asked the store to change the defective unit, which has happened.
After 10 days I get a new functional unit.

But I had the problem with the DJI CARE REFRESH plan associated with the defective unit.
I had to exchange too many messages with the DJI representatives so that the plan could be transferred to the new unit.
My recommendation is not to go blind, buy DJI CARE REFRESH only after you have activated the unit and you are sure it is working properly.
I misunderstood that the plan is only valid for units that have not been activated.
You have 48 hours to buy the plan.

Thanks for the assistance, but I'm still curious what was wrong with the old unit.  I probably will not find out. It was just a generator of emotions. There were some moments when I fully believed that I lost the unit. But I was lucky and came back.

Good day for you