Mavic Pro video weird vibrations

[airpett]

I was flying my new Mavic Pro the other day and suddenly the video started shaking weirdly as shown in the video linked below. I checked the gimbal and it was unobstructed (but I still had the little sticker with the tab, could this be an issue?). The rubber thing (vibration plate) was properly under all the three tabs, front and back. The temperature was quite low (-1 C) and it was pretty dark. Could these have caused the issues shown in the video? Or do you think there might be some other reason?

The IMU was calibrated on the day before on a flat and level floor and the compass just before flying. I also suspected the wind to cause this (3 m/s) but it didn't stop shaking by changing the place or flying lower. During the vibration, I tried to focus the camera multiple times, which took pretty long (I'm also thinking if the cold weather could have affected the focusing system). The vibration mainly happened while hovering the drone in place, not that much during moving around.

I also flew the drone today in a little warmer (+6 C) after doing all calibrations and the problem occured again, but not as badly as the other day. It was quite dark today as well, in case that matters.

To me the vibration looks like the gimbal motors would be turning it unintentionally, as the edges of the image move/shake instead of just getting waves on top of the image.

DETAILS:

Problem? Weird vibrations/shaking in shot video (live view and saved to SD card)

Was unit in a crash? no

What have you tried so far? Land the drone, take off again, re-calibrate IMU, compass and gimbal as well as fly with and without the gimbal cover (a little less vibration with the cover on, but problem still persists and the cover is not meant to be used airborne)

What device are you using? Nokia 8

What firmware are you running ( aircraft, remote controller)? newest (005)

What Go app version are you using? newest version from Play Store

Any modification? its all a stock unit

Did you change anything or install any apps? nope, nothing

Do you have a video or pictures of the problem? Yes, please see YouTube link of the original flight (problems start at around 0:43):

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

I can also upload today's footage if you need it, but will not yet since the connection upload is very slow.

I have also attached a DJI GO 4 FlightRecord from the flight in the above video. If you use AirData log viewer to see the map, the shaking started at the two West-most points while hovering there.

Thanks in advance, I'd very much appreciate any help!
If you have any questions, I'll answer ASAP to provide you with all the information that could help.

 

[Starz]

What was the reason for calibration of the imu and compass?
History on aircraft? Any previous crash issues?
Gimbal calibration?

 

[DanMan32]

Perhaps rubber mounts dislocated? Gimbal plate out of position?

 

[airpett]

What was the reason for calibration of the imu and compass?
History on aircraft? Any previous crash issues?
Gimbal calibration?

Hi! Thanks a lot for you reply!

I did the calibrations just as I thought it's best practice to calibrate them (especially the compass before every flight in a different place).

The drone is brand new for me, but it was bought from an outlet at the local elections store (might be old models or devices people have tried and returned). I know that myself I haven't crashed it, but can't be sure what has happened before I acquired it.

That is one of the things I'm trying to figure out, if this issue is probably due to a crash. I'd then take it back as they have a 50 day return policy + warranty even for the outlet products.

The gimbal was calibrated before flying and also before flying again today, and the issue still persisted.

 

[airpett]

Perhaps rubber mounts dislocated? Gimbal plate out of position?

Hi, thanks for your reply!

I already checked that the gimbal is properly fitted with all four rubber mounts in place, and the plate hasn't popped up from behind the tab.

 

[Starz]

Research info on calibration.
Gimbal calibration if there is an issue.
Imu, and compass information is in the app under the sensor section.
If calibration is needed, you will see it there. Otherwise, unnecessary.

 

[MAvic_South_Oz]

I think it's shivering with the cold.

Seriously, it's a very fine vibration.
Were the motors behaving normally, not revving harder for some reason at those points of time in the video ?

With the log the data might reveal something.
I think motor type issues might need the DAT file, can you get that file and upload ?
How to retrieve a V3.DAT from the tablet

Then it would be a job for @sar104 @BudWalker or others of that level of analysis expertise to assess.

 

[sar104]

There's no indication in the log file of any problems - no physical vibration. That looks to me like the camera image stabilization/focus is hunting.

 

[MAvic_South_Oz]

..
In that case . . .

I think it's shivering with the cold.

 

[airpett]

Research info on calibration.
Gimbal calibration if there is an issue.
Imu, and compass information is in the app under the sensor section.
If calibration is needed, you will see it there. Otherwise, unnecessary.

Ok, thanks for this information. I'll pay more attention to the sensor readings in the future to avoid unnecessary calibrations.

 

[airpett]

I think it's shivering with the cold.

Seriously, it's a very fine vibration.
Were the motors behaving normally, not revving harder for some reason at those points of time in the video ?

With the log the data might reveal something.
I think motor type issues might need the DAT file, can you get that file and upload ?
How to retrieve a V3.DAT from the tablet

Then it would be a job for @sar104 @BudWalker or others of that level of analysis expertise to assess.

Thanks a lot for your answer!

The cold is something I was suspecting too, if it like can't properly turn the gimbal or focus the camera due to it. I still have to confirm and also try if it happens in more day-lit conditions.

The motors didn't make any noise that would have stood out, but the drone was couples hundred meters away when the worst vibrations happened so I can't tell for sure.

The log data is attached below (DAT file renamed to TXT). I hope it's the right one since they had a little different timestamps and I landed and took off a couple times that time.

 

[airpett]

There's no indication in the log file of any problems - no physical vibration. That looks to me like the camera image stabilization/focus is hunting.

Thank you for looking into the logs! I also added the DAT file above in case it helps.

Do you think the cold would have caused this kind of focus/stabilization issues (based on your like)? Or some other reason for those?

I'll try to find some large indoor space with a higher temperature to test, since in Finland it's not likely to get any warmer this time of year.

 

[sar104]

Thanks a lot for your answer!

The cold is something I was suspecting too, if it like can't properly turn the gimbal or focus the camera due to it. I still have to confirm and also try if it happens in more day-lit conditions.

The motors didn't make any noise that would have stood out, but the drone was couples hundred meters away when the worst vibrations happened so I can't tell for sure.

The log data is attached below (DAT file renamed to TXT). I hope it's the right one since they had a little different timestamps and I landed and took off a couple times that time.

There is nothing unusual in the DAT file either. I don't think it had anything to do with the temperature - it appears that the image shake coincides with the autofocus system hunting for the correct focus, and I think the image stabilization is simply being confused by the low light levels, flashing lights and movement in the field of view.

 

[airpett]

There is nothing unusual in the DAT file either. I don't think it had anything to do with the temperature - it appears that the image shake coincides with the autofocus system hunting for the correct focus, and I think the image stabilization is simply being confused by the low light levels, flashing lights and movement in the field of view.

Thank you very much for your dedication in my issue!

I hope this to be the problem. I'll try to test it today in better lighting conditions between lectures at the University (in Finland it's currently enough light only from 9am to 2pm). I'll post the results as soon as I have performed the test.

 

[airpett]

Ok, I tested today in good light and the same issues happened again. It was seeking the focus and focusing took really long after tapping the focus point on screen. The shaking also happened using manual focus. I tried with and without the gimbal cover and there was dramatically less shaking with the cover on, but againg: That's not the way it's meant to be used.

I think it's possible that the person who returned it to the store has crashed it and that has damaged the focusing/stabilization system somehow. I have now tried flying on three or four days with different lighting and weather conditions, and the shaking is always there. I have double checked that the gimbal is properly in place and tried doing some sharp movements with the drone, which didn't cause this vibration. Its appearing is kind of random.

Thank you very much for everyone who has helped me with troubleshooting this issue!

 

[Starz]

Probably best to return and replace.

 

[FoxhallGH]

Maybe I'm wrong, but you seem to have fairly smooth video while the drone is moving, and the shaking appears to kick in when you get to the hover. Have a close look at the gimbal assembly and see if it's shaking when you hover just above head height (just above = don't get your head near the prop's! ). You can get a 'shake' in the body of a Mavic, due to the stabilisation GAIN settings being out (tis rare though!). When the Mavic is hovering, it's continuously compensating for movement in all axes and if GAIN is not quite set right, it can get into a situation where every movement it does to stabilise itself is slightly more than it needs, which means it then needs to go in the opposite direction, which then means it needs to go back again and you end up with a Mavic that is sitting in one place in the air, but it's shaking enough for the vibration to create a lower frequency harmonic shake in the gimbal assembly. If it's doing this, you can actually see the gimbal bouncing around while the Mavic is hovering above your head.
If you can't see gimbal shake while hovering - then I'm on completely the wrong track and ignore this ... However, if you are seeing the shake, have a look at this video that explains what causes it, and how to tune it out ... This is a critical adjustment, so if you do decide to change the GAIN settings, be sure to take a record of the settings before you make any changes!
[Note that this video describes this problem as it sometimes appears for Mavic Pro's using 'Platinum' low-noise prop's.] Let me know if it's relevant???

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[MAvic_South_Oz]

Even if stock like the OP says, with 8330 and not platinum 8331 props, there still could be some gain abnormality.

I would return it though, don't go stuffing around with an unmodified near new drone, return it and exchange if possible. . . \
Or, if too long after purchase but under warranty, open ticket and send to DJI per their directions.

 

[MercHornet]

I have a Mavic pro and mine had did the same thing when it’s cold. I noticed it made a unusual vibration when it did it . I ended up putting on the low noise props and I never had the weird vibration again. My Mavic did come with standard props.

 

[airpett]

Maybe I'm wrong, but you seem to have fairly smooth video while the drone is moving, and the shaking appears to kick in when you get to the hover. Have a close look at the gimbal assembly and see if it's shaking when you hover just above head height (just above = don't get your head near the prop's! ). You can get a 'shake' in the body of a Mavic, due to the stabilisation GAIN settings being out (tis rare though!). When the Mavic is hovering, it's continuously compensating for movement in all axes and if GAIN is not quite set right, it can get into a situation where every movement it does to stabilise itself is slightly more than it needs, which means it then needs to go in the opposite direction, which then means it needs to go back again and you end up with a Mavic that is sitting in one place in the air, but it's shaking enough for the vibration to create a lower frequency harmonic shake in the gimbal assembly. If it's doing this, you can actually see the gimbal bouncing around while the Mavic is hovering above your head.
If you can't see gimbal shake while hovering - then I'm on completely the wrong track and ignore this ... However, if you are seeing the shake, have a look at this video that explains what causes it, and how to tune it out ... This is a critical adjustment, so if you do decide to change the GAIN settings, be sure to take a record of the settings before you make any changes!
[Note that this video describes this problem as it sometimes appears for Mavic Pro's using 'Platinum' low-noise prop's.] Let me know if it's relevant???

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

Thanks for the reply and idea.

I tried to do what you suggested and flew the drone on front and on top of me for a period of time during which the vibration happened. I payed close attention to the gimbal and had a friend with me to have another pair of eyes, but neither of us could notice any visible vibration of the gimbal while the distortion happened in the video.

Because of that, I decided not to try playing around with the gain since it's hard to get rid of the non-visible vibration (in contrast to the shaking in the video you linked, where it could be pretty easily seen if the settings have an effect on the physical bouncing of 1cm up and down).

I'm gonna take it back for replacement and hope to get a properly working one (or even change to the Mavic Air with a better camera and more intelligent vision system).