Ever since updating to v01.00.0400 my once flawless MA has started glitching in yaw about 10-20 degrees to the left (very rapid sudden jerk while hovering) about a minute after take off.  This seems to be 100% repeatable.  The glitch does not seem to reoccur after the initial one but I am being very careful at this point and not venturing off to far.  Also I have noticed that the yaw is not "locked in" like it used to be.  The aircraft slowly turns left or right without any stick input while hovering.  I have re-calibrated the compass and the IMU and the sensor status page shows little to no interference with the sensors or the compass.

Any ideas?

I'm experiencing an incredible delay in PREVIEW footage (not in SD footage), as I've shown in videos on this link

video preview

i saw the same thing in my flight today.  the delay had me concerned.  I'm really hoping this is being addressed by DJI.  the footage is clean, but the delay caused me to oversteer and the video shows this.  Luckily, i have a good friend that does my post editing, so the footage isn't totally unusable.

ya I've had some issues with the gimbal slowly yawing with my Mavic pro

Hi !

Do your issue look exactly like in the movie at post 24# at this thread?

Please ensure that you flew your drone in a place without wind. Did you shoot a video at that time? Please provide us for further check, thanks.

Also see my post here.

DJI Elektra Posted at 2018-5-11 01:47
Please ensure that you flew your drone in a place without wind. Did you shoot a video at that time? Please provide us for further check, thanks.

No significant wind.

The glitch where the drone suddenly jerks in yaw is happening at least once per flight.

The slow turning when hovering happens all the time now.

Remember this started happening after the upgrade to v01.00.0400.  Before it was fine.

Logs are here:

https://drive.google.com/open?id=1W_OhfxTNpEnuEl7GuqBebgXA7tFIPTrN



DJI Flight Log Viewer - PhantomHelp.com which is associated with 042 DAT file.  Glitch at 2:22.6
DJI Flight Log Viewer - PhantomHelp.com which is associated with 043 DAT file.  Glitch at 0:48.8
DJI Flight Log Viewer - PhantomHelp.com which is associated with 044 DAT file.  Glitch at 0:44.3
DJI Flight Log Viewer - PhantomHelp.com which is associated with 045 DAT file.  Glitch at 1:15.7

asaw Posted at 2018-5-11 03:24
Also see my post here.

Not my issue.  I always check the compass sensor reading, the compass and the tablet agree, and that North on the App is close to where North truly is.

This glitch is not the same as the description in that post.  It is unrelated to any kind of rudder stick input.

the same thing happened to me, i took off and after it was 6 meters from the ground started to rotate by itself and drifting to the left crashing with a tree, i dont have video of the incident nor i have the logs because as soon as it happened i went to the store where i bought it to report the incident, it said it had good gps signal i also set a homepoint but the flight says no gps, its pretty weird the dji go app showed me too that i had good gps signal

here are some screenshots of the moment of the crash on the dji go app

This is one of the DAT files for one of my flights above (Flight 4),  The glitch is where the arrow is,  You can also see the slow drifting.  There is also a huge and various error between the IMU yaw and the magnetic yaw variables which I don't think is normal.

Flight 1 Glitch at 2:22.6


Flight 2 Glitch at 0:48.8


Flight 3 Glitch at 0:44.3


Flight 4 Glitch at 1:15.7

Hello, this is quite fascinating. Is it possible to fly without the controller ie: directly to phone or tablet? If this can be replicated without the controller than more fingers point to the drone/hardware/software. In the case that it can not be replicated the glitched may be caused by a sticking stick. The reason I ask this is I faced similar with another drone and couldn't put my finger on it until I tried another controller. If controller, it may be a simple fix if its a software issue and not a control contact. Cheers Anthony

ANTHONYVLOG Posted at 2018-5-11 14:16
Hello, this is quite fascinating. Is it possible to fly without the controller ie: directly to phone or tablet? If this can be replicated without the controller than more fingers point to the drone/hardware/software. In the case that it can not be replicated the glitched may be caused by a sticking stick. The reason I ask this is I faced similar with another drone and couldn't put my finger on it until I tried another controller. If controller, it may be a simple fix if its a software issue and not a control contact. Cheers Anthony

Can't be the stick because the log shows that the stick is in the neutral position during the glitch.  There is absolutely no stick input or stick input error in the DAT logs.

I can't fly it without the remote due to the fact that the drone DAT logs are encrypted and not usable by me for troubleshooting.  The DAT files I can access are on the controller/tablet.

Ok, learned a bunch today on how to get the right logs and decode them.

There seems to definitely be something wrong with my drone.  Either the latest firmware killed it or something is failing.

Since the prior mentioned flights, I have resinstalled the firmware with DJI Assistant, recalibrated the IMU, the compass and the gimbal.

I then flew 3 more flights this evening and here are the logs:
http://www.phantomhelp.com/LogViewer/Q5N8MYVTLGHXLOGK6D4C/
http://www.phantomhelp.com/LogViewer/9TSD6N0Z45M34CKPQ29H/
http://www.phantomhelp.com/LogViewer/YP5BVIDWU0Y1U8XNFXMF/
Here is a link to the .txt and .dat files off of the tablet:
Mavic Air Saga Day 2 - Google Drive

I experienced different kinds of glitches on all three of these flights.  Some yaw glitches like the ones described on the post above, some general glitches where the drone and the motors just make a sudden rapid glitchy sound and the drone shakes a bit, and the issue with the drone slowly turning in a hover.  I tried all kinds of movements with the rudder stick to gather different data.
At a few times I let the drone slowly rotate just to see how far it would go and I noticed it rotated so far one direction then started rotating back in the opposite direction.

Brantel Posted at 2018-5-11 17:41
Ok, learned a bunch today on how to get the right logs and decode them.

There seems to definitely be something wrong with my drone.  Either the latest firmware killed it or something is failing.

Hi Brantel,

I just submitted a support ticket for my MA to DJI for these EXACT same behaviours!

On May 5th when DJI released .400 firmware, I had experienced a complete IMU error and the drone went haywire and crashed, albeit with a little help from me as I still had altitude control and managed to make it go into a bush, and it suffered no physical damage. I then updated to .400 (via DJI Assistant 1.2.3) as soon as I got it home, and proceeded to do the following after the update:

• Reset MA to factory defaults
• Re-calibrate VPS using DJI Assistant
• Re-calibrate IMU on verified 0 degree flat surface x3
• Re-calibrate Compass in clear open field with no interference sources x3
• Re-calibrate RC via DJI GO 4 app on iOS
  

Even after all this, I still experienced the exact same behaviours as you described!

Perhaps it isn't a faulty drone, just faulty firmware! If that's the case, it really P___ES me off that DJI thinks we can do their beta testing for them like this, especially with a device that is airborne, and in the event of malfunctions like these, has the potential of causing property damage, or physical harm to innocent bystandards when this thing develops a mind of its own and decides to go crashing into something, or someone!

Brantel Posted at 2018-5-11 05:44
No significant wind.

The glitch where the drone suddenly jerks in yaw is happening at least once per flight.

Thanks for your update. I would help you forward the flight data. Please wait patiently.

EdisonW1979 Posted at 2018-5-11 20:25
Hi Brantel,

I just submitted a support ticket for my MA to DJI for these EXACT same behaviours!

At this point I am almost afraid to fly the drone because with the way it is glitching and rotating on its own, I am in fear of it flying off or dropping out of the sky.  My MA has been treated well since I got it.  Never even bumped it into anything or even had anything close to a minor crash.  It was working like a dream before the firmware update!

DJI Elektra Posted at 2018-5-12 01:28
Thanks for your update. I would help you forward the flight data. Please wait patiently.

Looking forward to your response.

DJI support suggested I reset factory defaults and reload the firmware using DJI assistant. Then recalibrate the IMU. (I did not expect this to fix the issue) I did this and for simplicity I flew one flight this morning to test it out. Same problems continue.

Full logs off the remote/tablet here:
Mavic Air Saga Day 3 - Google Drive

Txt log here:
DJI Flight Log Viewer - PhantomHelp.com

I pulled the DAT files off my phone of the last flight before I upgraded the firmware, notice how all the trends lay ontop of each other? This is something like I would expect out of a normal drone.

Then compare that to my flight this morning, notice how the YAW of each IMU does not agree and how the magnetic yaw is sorta floating around? The only thing that chanced was the firmware upgrade. (and a dozen recalibrations)

And then we have this other weird slow rotation going on. Here is a prime example of that where I let it go until it started rotating the other direction? Notice how it seemed to snap to the different IMU?

And here is yet another weird example of the slow rotation. In this one, it did not change direction but did go all the way to the other IMU... The IMU error seems to be in the opposite direction to the prior one so maybe that is why the drift behavior is different?

At any rate, due to the fact that there is only so much a person can do by resetting, recalibrating, testing etc. I have no choice but to attempt to work this out with DJI support. Unless someone else has any better advice? I suppose I could just wait for another firmware update and hope it fixes it but who knows....

I am convinced that the new firmware broke my perfectly good drone!

Brantel Posted at 2018-5-12 04:57
At this point I am almost afraid to fly the drone because with the way it is glitching and rotating on its own, I am in fear of it flying off or dropping out of the sky.  My MA has been treated well since I got it.  Never even bumped it into anything or even had anything close to a minor crash.  It was working like a dream before the firmware update!

I'm right there with you! My MA has been a dream to fly since I got it in April this year, even in deep canyons with less than 10 GPS satellite locks, it's always flown smooth and steady, giving me that sense of confidence, as it should.

Now, I have a severe sense of apprehension each time I fire up the motors, and it started with .400.

Looking at the additional data you just posted, it seems to prove there is something substantially wrong with the MA after this last update. One might think they'd be better off flying it in ATTI mode than trust the AI at this point.

Hoping DJI will action my support ticket soon.

Brantel Posted at 2018-5-12 08:51
I pulled the DAT files off my phone of the last flight before I upgraded the firmware, notice how all the trends lay ontop of each other? This is something like I would expect out of a normal drone.

What software are you using to view your DAT file data?

Is it something on PhantomHelp.com?

EdisonW1979 Posted at 2018-5-12 09:13
I'm right there with you! My MA has been a dream to fly since I got it in April this year, even in deep canyons with less than 10 GPS satellite locks, it's always flown smooth and steady, giving me that sense of confidence, as it should.

Now, I have a severe sense of apprehension each time I fire up the motors, and it started with .400.

I wish it were possible to manually switch to ATTI mode. Potentially it can help avoid crashes caused by other glitches in the autopilot. But in this case it seems like the IMUs has gotten upset by the FW update and since ATTI mode relies heavily on IMUs, switching to it won't make much difference.

EdisonW1979 Posted at 2018-5-12 09:37
What software are you using to view your DAT file data?

Is it something on PhantomHelp.com?

I use a combination of tools:

The logs that we can look at are stored on the mobile device connected to the remote controller.

My go to apps
CscView is the bomb!  It can open all the MA's logs except the ones that are stored in the Drone.  DJI encrypted those so nobody but them can open them.

I use this Phantom Help Log viewer as well
I use this if I need to make a CSV file out of the TXT file log.

Nothing I have tried works to solve the issue.  I am working with DJI support.  Just so folks don't assume this is an isolated event, I have been capturing comments of others as I see them regarding the same issue...The list is growing!  Names blurred out for privacy.

Folks,

Progress is being made in identifying the root cause of these issues.  A very promising post was made to one of my post on Face Book from a DJI support group manager.  The post states that they have been able to replicate the issue on their end and that a fix for it is imminent.

This is good news and praise for DJI's support group for listening!

Brantel Posted at 2018-5-14 08:59
Folks,

Progress is being made in identifying the root cause of these issues.  A very promising post was made to one of my post on Face Book from a DJI support group manager.  The post states that they have been able to replicate the issue on their end and that a fix for it is imminent.

That is AWESOME news! So far I haven't gotten very far in my official support case with DJI, so am very hopeful they'll just put out a SW update and get all our MA's back in full flying form!

Hi, Friends!
Ever since updating to v01.00.0400 my Mavic Air do not compas calibrate..... also other problems appeared during flight. How can I recover previous version of an update? Are there any recomendations from DJI?

Hi guys!
I heard a lot of complaints regarding the latest firmware update. I contacted DJI customer service and they told me they are aware of the issue and try to solve it as soon as possible ( doesn't sound promising!) Also, they told me that not all of the Mavic Air drones behave unusual after this update, but they are not able to tell me which batch or serial number is involved !

My drone is brand new and the only test i did, was just indoor, after this... " amazing" firmware update! I couldn't find time and a proper place in London to carry on more extensive tests.

In my indoor tests, I flied around 40 min, recording, moving up and down and rotating ( yaw). I moved also the gimbal up and down, while yaw. I didn't notice any obvious issue !

I get a feeling of a slight drifting while hoovering ( maybe few centimeters), maybe a slight rotation ( few degrees in horizontal plane) - but I assumed it is " normal" as it was in a small environment ( a room) close to large objects ( which could reflect the air back to the propellers) and create incremental moves. Plus...it is a drone, not a poster to stay still !

I don't have a reference point to compare how my Mavic Air is behaving after the latest firmware update, comparing with how should behave.
Can anyone help with some information regarding " the ideal " behavior of Mavic Air, the one described by many to be present before this " infamous update" ?
This micro-movements while hovering it is the only thing I noticed, but I can't compare with how it's supposed to be, and as I said, I didn't make any extensive test yet. Are they normal ?
Thank you for your time and your help!

So...

The rumours I've heard about DJI support being deplorable from other forum members here was well justified...

After sending them nearly 2GB of logging data directly off my drone, pointing them to this forum thread for cross referencing data, their support rep kept providing responses which seemed like he was reading off a script, didn't want to put me in touch with what he called their "Service team" to go over the logs, didn't want to acknowledge the software issues, and insisted they won't do anything until they get my drone for "repairs", when there was no damage. They couldn't even tell I was in Canada, thinking I was in the US!

If DJI is considered the best drone company, or the Apple of drones, they SERIOUSLY need to work on their support, as it's the polar opposite of what it should be. I'm frankly disgusted by the support I received, that a company who sells products this expensive, with this much potential to cause damage and/or injury, has such poor support.

It's making me regret buying anything with DJI's logo on it...

Hopefully one of the DJI admins here will agree to take my logs directly to the engineers and either tell me what they find, or at least acknowledge whether the issues I've been experiencing are the same as Brantel's, and being related to the latest .0400 firmware.

Elysian Posted at 2018-5-15 09:26
Hi guys!
I heard a lot of complaints regarding the latest firmware update. I contacted DJI customer service and they told me they are aware of the issue and try to solve it as soon as possible ( doesn't sound promising!) Also, they told me that not all of the Mavic Air drones behave unusual after this update, but they are not able to tell me which batch or serial number is involved !

What you are describing would be typical inside. Inside the drone is using the bottom vision sensors to do most of its position control so this issue would likely not show up inside. Even outside typically you have to get up high enough that the bottom sensors are not seeing the ground for the issues to really become noticeable.

楼主你好，我们会进一步优化这个问题，同时建议你尽量去空旷地带起飞。

Hello together,

I have the same problem ONLY since 0400 update!
slowly turing in yaw to the left side, but ONLY if I paned before! If i rais straightly up
the air do not move in yaw it stands still.
I remember to have had that same problem with an earlier Air  model, this was corrected by firmware,
now it seems to be back with 0.400...so DJI will knows how to solve that problem,
because they solved the same problem earlier.

I have the same problem with drifting MA after update to 0400.