fans46aaea05 Posted at 2017-2-21 06:27
Hi everyone,

I am admittedly a novice pilot but my Mavic did the same thing too this weekend. I'm not ruling out some kind of pilot error, but in my case i had good GPS strength, auto took off (habit) and flew for about 2 minutes when it suddenly switched to atti mode and drifted off to the right at full speed. There was some wind for sure but not excessive. Mavic did not respond to any commands and ended up hitting a tree.

First I think you should start a new thread, more people will see it and may be able to help.

Second you should sync your flight log to dji cloud.

Third you should if you can post your flight logs in new thread you can upload your logs to http://www.phantomhelp.com/LogViewer/Upload/ and download the link to your new thread.

You don't have to calibrate compass or IMU before each flight, when these need calibration you will be prompted.

Ejder this sudden loss of GPS signal and the uncontrollably drifting is very worrisome. As far as I know DJI has not commented on this catastrofic problem that causes a lot of trouble and expences to us Mavic owners. Its a damn shame and is a warranty issue no doubt. But the support centres  are very clever to blame the pilot, I know.

I have to add to my posting above that its obvious that the Mavic is a premature born child. I rather be out flying. Instead I have to sit and read of all the problems and fimware issues. The drone should preform as the manual says , period. Can you imagine having a car or a TV with this ongoing problems ? Yes I finally understand that we are all Beta-testers. The Mavic owners seems to be the most patient customers in the world. I'm not.

fansc2962bbc Posted at 2017-2-21 07:44
I have to add to my posting above that its obvious that the Mavic is a premature born child. I rather be out flying. Instead I have to sit and read of all the problems and fimware issues. The drone should preform as the manual says , period. Can you imagine having a car or a TV with this ongoing problems ? Yes I finally understand that we are all Beta-testers. The Mavic owners seems to be the most patient customers in the world. I'm not.

You are completely right, i got tired with explaining this problem to support team. So, i already put my mavic and 1500 USD to trash and keep flying with my P3A. It is fact that we are Beta-users.

z4k4tt4ck Posted at 2017-1-26 01:29
WRONG .... There is never no wind .... Never .... Never. Think.

WOW, You're a little wrong yourself there mate.  
for a start, 800km into the Australian desert in the middle of a summers day (47c) i can guarantee there is no wind 8 feet off the ground and there was drift with my p4p in atti mode. about a half meter a second.
spare a thought for other peoples situations before spewing hate.

gibbd Posted at 2017-1-26 02:05
This actually made me laugh.   How can you say there was no wind.   No wind situations exist in about 0.01% of the time (or indoors).

If you are indoors and blow at your mavic in ATTI mode.... it will drift.  Ive tried it!

giddb,

I'm sorry to say and see a need to correct you.
plenty of satellites AND strong satellite signal (which are two different things).
With the Mavic system the satellite's that are picked up and locked in.
Satellites that are not picked up do not register and you never know that they are there.
So with that said the Number and strength of the satellite's registering with the Mavic
are the same.
The only time I know the Number of satellite's and satellite signal strength would Differ is
with a handheld GPS it shows all present satellite's above you But the GPS only picks out the
one's it needs to work properly are the one's with the strongest signals
This is a tricky question like which came first? the chicken or the egg?
But for the Mavic system you only receive and register the one's it locks onto
The system does not pick up 12 satellite's and then show's it only uses 3
Fly Safe -- Fly Free  
  

Yes I had this happen twice and both times uncontrolable but had enough height and remaining battery to get it home but the lates firmware seems to have solved this by reducing the power as its deffinatly slower and I often have to use sport mode to fight wind

Reading the OP reminds me of a guy learning to fly helis a few years ago . He would get it into a hover and do ok then fly out a bit and then back it in towards him to land . One day a friend and I noticed him getting it into a hover then in just a minute he would have to set it down real hard and start cursing about the dang thing won't fly right .  After about the 5th time of him setting it down real hard and almost crashing we walked over . He said he'd been flying this thing for a couple months just fine but now it didn't respond to his stick inputs the way it should sometimes . My friend ask if he had ever flown RC planes or anything else and he said no , he was starting on helicopters . He asked my friend to try and fly it and tell him what was going on . My friend checked the heli and radio over then put it into a nice hover and before the guy could say anything he took off like a bat outta hell into the wild blue yonder and commenced to doing a few figure 8's , piros etc , nothing real hard but it certainly showed there was no control problem . He bought it in for a landing nose first and set it down with a short skid slide stopping about 10' in front of us . The guy had a slight surprised expression on his face but said that's cool , that's what I'm trying to learn, to bring it toward me like that . My friend said yeah I noticed every time you almost crashed you were nose in hovering . You do know the controls are backwards when it's facing you right ?

rydfree41 Posted at 2017-4-9 08:13
Reading the OP reminds me of a guy learning to fly helis a few years ago . He would get it into a hover and do ok then fly out a bit and then back it in towards him to land . One day a friend and I noticed him getting it into a hover then in just a minute he would have to set it down real hard and start cursing about the dang thing won't fly right .  After about the 5th time of him setting it down real hard and almost crashing we walked over . He said he'd been flying this thing for a couple months just fine but now it didn't respond to his stick inputs the way it should sometimes . My friend ask if he had ever flown RC planes or anything else and he said no , he was starting on helicopters . He asked my friend to try and fly it and tell him what was going on . My friend checked the heli and radio over then put it into a nice hover and before the guy could say anything he took off like a bat outta hell into the wild blue yonder and commenced to doing a few figure 8's , piros etc , nothing real hard but it certainly showed there was no control problem . He bought it in for a landing nose first and set it down with a short skid slide stopping about 10' in front of us . The guy had a slight surprised expression on his face but said that's cool , that's what I'm trying to learn, to bring it toward me like that . My friend said yeah I noticed every time you almost crashed you were nose in hovering . You do know the controls are backwards when it's facing you right ?

You do know the controls are backwards when it's facing you right?

Yikes. Hopefully the owner never travels to a country where he'd be expected to drive on the opposite side of the road from what he's used to.

DroneFlying Posted at 2017-4-9 08:49
You do know the controls are backwards when it's facing you right?

Yikes. Hopefully the owner never travels to a country where he'd be expected to drive on the opposite side of the road from what he's used to.

LOL, that would actually be easier though than having the car go left when you steered right .

Don't know if still the same in newer firmwares, but have had this happen to me before..

What I found in testing later, is ATTI mode DRASTICALLY reduced input form the controller when it switched to that mode (I assume to make input smoother for the user when the freak out and realize they are now solely in control lol).

Believe it or not, "pumping" the stick all the way to the edge and back repeatedly in the direction of choice, after multiple times would start to increase the input until back to normal (again obviously a programmed behavior, it was very replicatable inside evey time, and could even easily notice the input increasing steadily on each "pump".

Maybe some of this still exists in the newer firmwares, and your issue is your input was practically ignored when ATTI mode kicked in..

(happened to me 3 times in total before)

Ejder Bingül Posted at 2017-2-24 11:59
You are completely right, i got tired with explaining this problem to support team. So, i already put my mavic and 1500 USD to trash and keep flying with my P3A. It is fact that we are Beta-users.

Amen to that!!

Phantom 4, flew out of the box, NEVER even ONE kind of glitch, never ever went to Phantom 4 forum.

I now live in the Mavic Forum, and have had a long list of issues that come and go from firmware to firmware.

Remind me EXACTLY like when had Parrot Bebop.. Issues.. Firmware fixed them, creating more Issues.. Rinse-Repeat!!


Ugh!

SnowBirdNL Posted at 2017-4-10 04:37
Amen to that!!

Phantom 4, flew out of the box, NEVER even ONE kind of glitch, never ever went to Phantom 4 forum.

Having similar issues with my mavic - used the litchi app and ran a waypoint map, it lost video signal a few times at the farthest  distance but came back. 3/4 on the way through the camera started to drift and the recording started and stopped server times. It completed the Litchi program and stayed at hover like it should.  I couldn't fly it back to me was yawing to the right and stoping over and over again going in circles.  I closed the litchi app and open DJI go 4 - here is a link of the
Also here is a info from my  Mission

mfpetrow1 Posted at 2017-4-23 05:10
Having similar issues with my mavic - used the litchi app and ran a waypoint map, it lost video signal a few times at the farthest  distance but came back. 3/4 on the way through the camera started to drift and the recording started and stopped server times. It completed the Litchi program and stayed at hover like it should.  I couldn't fly it back to me was yawing to the right and stoping over and over again going in circles.  I closed the litchi app and open DJI go 4 - here is a link of the video
Also here is a info from my  Mission

Well I just had the exact same thing happen to me tonight. My controller was showing gps mode with lots of satellite then all of a sudden ATTI mode kicked in. I had no control what so ever. This has happened on all of my flights so far. Wide open areas. I feel it has to do with the latest firmware update. I will not fly my mavic until this is fixed. I had no control of the drone at all. Very difficult to land.

I had the exact same issue indoors, ;ed to my first crash.  Bottom line, don't fly indoors...if it goes to atti mode good luck.

I want to say that I had a similar experience myself! My Mavic was following a few successful flights, and having experienced a crush!At first, I did not understand what had happened, after receiving several warnings, and Mavic Turned to the back in a fraction of a second and fell to the branches of the Pine tree!
I am actually very happy about that tree because the damage that came up is very small. After I turned to the seller and explained what happened, I got the info that I had to open Log on Dji Go APP and send it to Dji cloud! I soon realized that the crush of the drone was caused by a silent killer called switching gps to atti mode! The dron is now in the service test and I'm waiting for feedback whether she is absolutely safe for the flight! Of course, early on, DJI claims it's not a problem with firmwire! The first problem is that the dron turned to the back that was determined to have occurred at a time when it was switched from one mode to another mode, And the other problem (by me) is that the drone begins to lose control by the pilot that manages it (many users report)I hope that people from DJI have some better explanation than the fact that the reason is breeze or i do not know what, but the fact is that many cases end up with the fall of the drone and the big damages.
As the third thing I do not understand is how on a meadow where there is no building can the loss of GPS signal occur??? DJI always blames us pilots, I have been flying the RC model's for more than 5 years. I control the models without the autopilot and I will never let it fall even if there is a wind. I personally consider that there is a bug, which is why the dron loses control and DJI for obvious reasons deny it (My personal opinion)

Matt77 Posted at 2017-4-30 05:00
I want to say that I had a similar experience myself! My Mavic was following a few successful flights, and having experienced a crush!At first, I did not understand what had happened, after receiving several warnings, and Mavic Turned to the back in a fraction of a second and fell to the branches of the Pine tree!
I am actually very happy about that tree because the damage that came up is very small. After I turned to the seller and explained what happened, I got the info that I had to open Log on Dji Go APP and send it to Dji cloud! I soon realized that the crush of the drone was caused by a silent killer called switching gps to atti mode! The dron is now in the service test and I'm waiting for feedback whether she is absolutely safe for the flight! Of course, early on, DJI claims it's not a problem with firmwire! The first problem is that the dron turned to the back that was determined to have occurred at a time when it was switched from one mode to another mode, And the other problem (by me) is that the drone begins to lose control by the pilot that manages it (many users report)I hope that people from DJI have some better explanation than the fact that the reason is breeze or i do not know what, but the fact is that many cases end up with the fall of the drone and the big damages.
As the third thing I do not understand is how on a meadow where there is no building can the loss of GPS signal occur??? DJI always blames us pilots, I have been flying the RC model's for more than 5 years. I control the models without the autopilot and I will never let it fall even if there is a wind. I personally consider that there is a bug, which is why the dron loses control and DJI for obvious reasons deny it (My personal opinion)

If you upload the flight log from your mobile device using the instructions here and provide a link to it we may be able to provide some useful information on what might have gone wrong.

Matt77 Posted at 2017-4-30 05:00
I want to say that I had a similar experience myself! My Mavic was following a few successful flights, and having experienced a crush!At first, I did not understand what had happened, after receiving several warnings, and Mavic Turned to the back in a fraction of a second and fell to the branches of the Pine tree!
I am actually very happy about that tree because the damage that came up is very small. After I turned to the seller and explained what happened, I got the info that I had to open Log on Dji Go APP and send it to Dji cloud! I soon realized that the crush of the drone was caused by a silent killer called switching gps to atti mode! The dron is now in the service test and I'm waiting for feedback whether she is absolutely safe for the flight! Of course, early on, DJI claims it's not a problem with firmwire! The first problem is that the dron turned to the back that was determined to have occurred at a time when it was switched from one mode to another mode, And the other problem (by me) is that the drone begins to lose control by the pilot that manages it (many users report)I hope that people from DJI have some better explanation than the fact that the reason is breeze or i do not know what, but the fact is that many cases end up with the fall of the drone and the big damages.
As the third thing I do not understand is how on a meadow where there is no building can the loss of GPS signal occur??? DJI always blames us pilots, I have been flying the RC model's for more than 5 years. I control the models without the autopilot and I will never let it fall even if there is a wind. I personally consider that there is a bug, which is why the dron loses control and DJI for obvious reasons deny it (My personal opinion)

Hi matt77, when you loose gps and go into Atti mode your AC should hold its height, although it will drift on the wind this will be exaggerated by your movements on the sticks, it should not drop its height.

With regards to flying in an open field, of course you can loose gps for many reasons cloud cover is just one,
Hope you back flying soon.

Can you upload the flight log so we all can have a look at it and try to help out. Maybe some of the guys around will be able to notice something that might explain what happened. Also, the .DAT file would be nice. What has been DJI response after looking at your flight log?

Cheers!

Hello Ejder,

I have had a similar experience, a couple of days ago with Mavic Pro, Indoor (factory) around 3-4 m of the ground. ( Atti mode ); and had a similar stresful experience.

What caused the problem is light breeze, and to overcome that using Right Stick ( @ Mode 2 ) solved most of the drifting issues, BUT I think the flight mode is the critical thing.

I tried running into very light breeze in tripod mod, and I have struggled to make it move in counter direction, probably because in tripod flight mode ( ATTI ) the rotor speeds does not deviate to a state where quadrotor can get to greater angle so that it can overcome and accelerate in counter direction to wind.

I will try to switch to sport mode quickly when this happens again and I will post the updates.

SnowBirdNL Posted at 2017-4-10 04:34
Don't know if still the same in newer firmwares, but have had this happen to me before..

What I found in testing later, is ATTI mode DRASTICALLY reduced input form the controller when it switched to that mode (I assume to make input smoother for the user when the freak out and realize they are now solely in control lol).

Sorry for quoting your old post but WOW, if I recall correctly this wasn't mentioned anywhere in the manual, firmware releases or at least in the warnings (ATTI mode DRASTICALLY reduced input).
Can DJI confirm this? I was under the impression that ok I've no GPS positioning and I should countereact normally as one would, not ramming the mode 2 right stick 100% a few times just to tame the Mavic.

I also posted in this thread here requesting for precision inputs whilst in Atti mode.

Nazar78 Posted at 2017-7-26 10:44
Sorry for quoting your old post but WOW, if I recall correctly this wasn't mentioned anywhere in the manual, firmware releases or at least in the warnings (ATTI mode DRASTICALLY reduced input).
Can DJI confirm this? I was under the impression that ok I've no GPS positioning and I should countereact normally as one would, not ramming the mode 2 right stick 100% a few times just to tame the Mavic.

"What I found in testing later, is ATTI mode DRASTICALLY reduced input form the controller when it switched to that mode"

WOW, if I recall correctly this wasn't mentioned anywhere in the manual, firmware releases or

It wasn't mentioned because it's not correct. I've flown in ATTI mode and the Mavic functions just fine. If the person you replied to experienced ATTI mode it was most likely triggered by some problem such as taking off from a geomagnetically distorted location, which can indeed result in difficulty controlling the aircraft. However, that's a separate issue and not a problem inherent to flying the Mavic in ATTI mode.

I ran into the same problem when i first bought my Mavic back in January I did all the proper step in updating software and calibration of IMU, with no success. and at the end i ended up sending it back to DJI for repairs using my DJI care plan in turns it was replace with a new one.

DroneFlying Posted at 2017-7-26 13:08
"What I found in testing later, is ATTI mode DRASTICALLY reduced input form the controller when it switched to that mode"

WOW, if I recall correctly this wasn't mentioned anywhere in the manual, firmware releases or

Hi DroneFlying,

Noted. Just to clarify, my concern now is not with the Mavic switching into Atti mode as I'm well aware it can happen anywhere anytime. The real concern is unresponsive inputs while in Atti mode.

Nazar78 Posted at 2017-7-27 01:03
Hi DroneFlying,

Noted. Just to clarify, my concern now is not with the Mavic switching into Atti mode as I'm well aware it can happen anywhere anytime. The real concern is unresponsive inputs while in Atti mode.

I understand, but it's important to clarify what you mean by "ATTI mode"; if you mean what's reported as "ATTI" when you take off from a geomagnetically distorted location then that isn't the same as true ATTI mode in which the Mavic flies just fine.

https://forum.dji.com/thread-105281-1-1.html
I did the same thing.
Could it be a manufacturing defect or software / firmware? what do you think?
I found compass error, imu error, and in the log I find yaw error.
The unpleasant thing is that I did not manage to find the drone because it was directed into a dense forest, and the assistance after 1 week of work has not yet answered me

DroneFlying Posted at 2017-7-27 02:06
I understand, but it's important to clarify what you mean by "ATTI mode"; if you mean what's reported as "ATTI" when you take off from a geomagnetically distorted location then that isn't the same as true ATTI mode in which the Mavic flies just fine.

I understood Atti mode which is hold altitude using its sensors but if there's geomagnetically distortion during take off then it's not true Atti as the sensors are off.

What I mean is despite true Atti or not (maybe IMU gone haywire during take off which is not in this case), we need full precise responsive control of the aircraft which you said in Atti mode Mavic flies just fine but not in my case as it does not respond to me flicking the forward stick few times to stop the backward drift.

Sorry I'm still new to this forum but I'm not sure if you've hovered an aircraft without all these advanced sensors before. A quick flick on the elevator or aileron of the said drifting aircraft, it will respond with a nod to the direction you flick even the slightest or even in the wind. That said, I would expect in Atti mode, Mavic will try to hold its altitude (despite sensors are off or not) but still respond to my sticks yaw/elevator/aileron "small" corrections which it does not.

Nazar78 Posted at 2017-7-27 05:21
I understood Atti mode which is hold altitude using its sensors but if there's geomagnetically distortion during take off then it's not true Atti as the sensors are off.

What I mean is despite true Atti or not (maybe IMU gone haywire during take off which is not in this case), we need full precise responsive control of the aircraft which you said in Atti mode Mavic flies just fine but not in my case as it does not respond to me flicking the forward stick few times to stop the backward drift.

So to make sure I'm clear on what we're talking about, you're referring to true ATTI mode that can occur when flying indoors with no (or at least very weak) GPS reception, correct? Because in your other recent post you specifically mentioned that you're talking about flying indoors, which would presumably be true ATTI mode. And if I understand correctly, you're saying that, "it does not respond to me flicking the forward stick few times to stop the backward drift".

If that's the case then yes, I'd like to see the DAT file from that, though I have to admit that I don't have much experience flying a Mavic indoors or analyzing DAT files from those who have. Am I correct in understanding that you not only were flying indoors but did so with VPS disabled? Because flying without GPS but with VPS enabled results in OPTI mode and not ATTI.

DroneFlying Posted at 2017-7-27 06:03
So to make sure I'm clear on what we're talking about, you're referring to true ATTI mode that can occur when flying indoors with no (or at least very weak) GPS reception, correct? Because in your other recent post you specifically mentioned that you're talking about flying indoors, which would presumably be true ATTI mode. And if I understand correctly, you're saying that, "it does not respond to me flicking the forward stick few times to stop the backward drift".

If that's the case then yes, I'd like to see the DAT file from that, though I have to admit that I don't have much experience flying a Mavic indoors or analyzing DAT files from those who have. Am I correct in understanding that you not only were flying indoors but did so with VPS disabled? Because flying without GPS but with VPS enabled results in OPTI mode and not ATTI.

Yes you got me right. Indoors, Atti mode, VPS disabled due to low light condition and the controls are unresponsive.

Ok about retrieving the DAT file, I've followed the instructions from the link you gave previously, but I see no list in the "Flight Data" from the DJI Assistant 2 v1.1.2-2 on Win10 (refer attached screenshot). Instead it asked to click on the SD card icon which is nowhere to be found on the interface either. Everything is the latest, firmware, app, software etc.


Wherelse in the Android DJI Go 4 app I see two flights data (refer attached screenshot) and I'm not sure how to export this?


A little OT, I just finished playing around with the simulator on the Android DJI Go 4 app (latest version), oh man, I don't know if the sim is lagging or I'm useless or what? Full right yaw, took couple of secs sometimes longer before it starts turning! Same goes to the rest of the stick controls and other interfaces on the screen. Frustrated then I downloaded this Sky Viper Flight Simulator from Google Play store and I'm doing F8s, nose-ins on the first launch with no stress at all... Updated, ok I think DJI Go 4 Android app is buggy, because I tried the sim from DJI Assistant 2 the sticks response is what I'm expected if ias though t's in Atti mode but eventually after few minutes the model freeze on PC screen lol.

Nazar78 Posted at 2017-7-27 10:33
Yes you got me right. Indoors, Atti mode, VPS disabled due to low light condition and the controls are unresponsive.

Ok about retrieving the DAT file, I've followed the instructions from the link you gave previously, but I see no list in the "Flight Data" from the DJI Assistant 2 v1.1.2-2 on Win10 (refer attached screenshot). Instead it asked to click on the SD card icon which is nowhere to be found on the interface either. Everything is the latest, firmware, app, software etc.

I see no list in the "Flight Data"

The tab with the DAT files is the one that's now called "Data Upload", so you should find it there.

As far as the simulator goes, I've never used it so I couldn't say but yes, from what I hear the Android app seems to have more problems in general than the iOS version.

DroneFlying Posted at 2017-7-27 10:36
I see no list in the "Flight Data"

The tab with the DAT files is the one that's now called "Data Upload", so you should find it there.

Ok I went to the "Data Upload", it gave a prompt then I continue and the small grey circle is still spinning (loading or stall?) since you last posted. Have tried rebooting the MP few times still no go. However, when I power on the MP without DJI Assistant 2 running, there's a Fat32 drive containing many files mostly DATs.

Since I'm not sure which of the four files from the 23rd July, I've zipped all four of them and uploaded it here. In the DJI Go 4 app it shows two flights, there's no timestamp though, but after the unresponsive drifting backwards incident, I tried rebooting and firing the motors again this time without the props hence two flights recorded.

Thanks again Sir for helping to look into this.

Nazar78 Posted at 2017-7-27 11:25
Ok I went to the "Data Upload", it gave a prompt then I continue and the small grey circle is still spinning (loading or stall?) since you last posted. Have tried rebooting the MP few times still no go. However, when I power on the MP without DJI Assistant 2 running, there's a Fat32 drive containing many files mostly DATs.

Since I'm not sure which of the four files from the 23rd July, I've zipped all four of them and uploaded it here. In the DJI Go 4 app it shows two flights, there's no timestamp though, but after the unresponsive drifting backwards incident, I tried rebooting and firing the motors again this time without the props hence two flights recorded.

I looked at your DAT files and FLY016 seems to be the relevant one, but I didn't really notice anything unusual in it. Other than starting and stopping the motors multiple times (sometimes, as you mentioned, without propellers) I didn't see a lot of control stick input, though that's not surprising considering that you were indoors. When you mentioned that the Mavic seemed unresponsive, I'm guessing that you were referring to the two taps on the throttle that appear below as the spikes in the gold plot at around 106 and 108. As you can see, the motor speeds (in purple) did briefly increase at those times, but then returned to their original speeds as soon as you released the throttle.

You may also notice in the graph that there's a bit of a delay between your taps on the throttle and the motor speed increases, but that's normal too. That's because your Mavic was close enough to the floor that it considered itself to still be in the process of taking off (indicated by the blue background shading), and when it's taking off or landing the controls seem sluggish due to muted response to stick input. People normally only notice that during landing when they try to fine-tune the location where their Mavic touches down, but apparently it responds the same way during takeoff.

The bottom line is that I believe once you're flying outside and are sufficiently high above the ground that you won't see any problem with the controls being unresponsive.

DroneFlying Posted at 2017-7-28 07:40
I looked at your DAT files and FLY016 seems to be the relevant one, but I didn't really notice anything unusual in it. Other than starting and stopping the motors multiple times (sometimes, as you mentioned, without propellers) I didn't see a lot of control stick input, though that's not surprising considering that you were indoors. When you mentioned that the Mavic seemed unresponsive, I'm guessing that you were referring to the two taps on the throttle that appear below as the spikes in the gold plot at around 106 and 108. As you can see, the motor speeds (in purple) did briefly increase at those times, but then returned to their original speeds as soon as you released the throttle.

You may also notice in the graph that there's a bit of a delay between your taps on the throttle and the motor speed increases, but that's normal too. That's because your Mavic was close enough to the floor that it considered itself to still be in the process of taking off (indicated by the blue background shading), and when it's taking off or landing the controls seem sluggish due to muted response to stick input. People normally only notice that during landing when they try to fine-tune the location where their Mavic touches down, but apparently it responds the same way during takeoff.

Hello DroneFlying,

Really appreciate you looking into this. I'm planning to fly it outside this weekend for the first time after 1 week purchasing it. Hoping it won't go into Atti mode just to find out if the sticks are responsive.

"When you mentioned that the Mavic seemed unresponsive, I'm guessing that you were referring to the two taps on the throttle that appear below as the spikes in the gold plot at around 106 and 108." <-- Actually I'm referring to unresponsive back elevator while I was trying to stop the Mavic from drifting backwards.
I've just tried analyzing the DAT files with CsvView myself but can't find the matching flight pattern of my very first hover from the 23rd July, guess I have lots to learn. Are the DAT files stored in the Mavic timestamped in UTC?

Corrections: Actually I'm referring to unresponsive forward elevator while I was trying to stop the Mavic from drifting backwards.

Nazar78 Posted at 2017-7-28 11:47
Corrections: Actually I'm referring to unresponsive forward elevator while I was trying to stop the Mavic from drifting backwards.

Well, again, I didn't really notice any input like that during the actual flight (not counting the times you started it up without props), so are you sure you posted the right DAT file? I'd suggest that you go ahead and fly it outside and I think you'll find that it responds just fine to the controls; the likelihood of you actually encountering ATTI mode outdoors is very low anyway.

DroneFlying Posted at 2017-7-28 13:05
Well, again, I didn't really notice any input like that during the actual flight (not counting the times you started it up without props), so are you sure you posted the right DAT file? I'd suggest that you go ahead and fly it outside and I think you'll find that it responds just fine to the controls; the likelihood of you actually encountering ATTI mode outdoors is very low anyway.

That's the thing, I don't even know which is the correct DAT haha. The timestamps doesn't seems to match, I was hovering in late evening but all the time stamps are at night. Oh well see how it goes this weekend. Thanks again!

DroneFlying Posted at 2017-7-28 13:05
Well, again, I didn't really notice any input like that during the actual flight (not counting the times you started it up without props), so are you sure you posted the right DAT file? I'd suggest that you go ahead and fly it outside and I think you'll find that it responds just fine to the controls; the likelihood of you actually encountering ATTI mode outdoors is very low anyway.

Hi DroneFlying,

Just to update, took your advice, hold my breath and brought it outside to a small field for the first time (finally). Tried few f8s it flew beautifully in the wind and I think I'm prepared to handle it if touch wood it goes into Atti.

Again, many thanks a lot for your inputs and advice!

Nazar78 Posted at 2017-8-5 10:04
Hi DroneFlying,

Just to update, took your advice, hold my breath and brought it outside to a small field for the first time (finally). Tried few f8s it flew beautifully in the wind and I think I'm prepared to handle it if touch wood it goes into Atti.

Thanks for the update; a field was a good choice and I'm glad to hear that it went well. Caution is a good thing when it comes to flying, and I'm sure you'll continue to do fine.

My Phantom 4 crashed because it could not hover steadily when taking off without or with low GPS signal. I did not set it to ATTI mode nor disable the sensor manually. And, it was showing "ready to go" on DJI app.

DJI Support replied that "the incident was not caused by any product malfunction factors" because it was in ATTI mode (which I did not set) :
"1. ATTI is one status of the aircraft. In that mode, the aircraft was not able to hover, skid or avoid obstacle, because there was no GPS signal and also the obstacle sensor system is not available, which is the feature.
It is not that someone manually disable the obstacle sensor, but when there is no GPS signal and also the obstacle sensor system is not available, the aircraft would enter the ATTI mode.
2. We never informed that the pilot disable the obstacle sensor in this case. But when the condition for the obstacle sensor to work, is not met, the sensor won't work."

I personally think that the sensor obstacle sensor should be switched on under any circumstances since it is the core feature of Phantom 4 to avoid accident as much as possible. Unless the obstacle is out of the sensor detection limitation. Also, DJI app should not allow any DJI drone to fly off without the aforementioned sensor being switched on at least.

Started manually today in the P-mode.
500 meter out of a see the equipment switched to anti mode.
It was to far away to see the P4P.
But I  had still video in the receiver and could therefore rescue the copter and steer back and after some 100 feet I got GPS-connection agin.
Latest firmware updated weeks before.