My Spark since day 1 has always move side to side while hovering. i have taken off the propellers and place them again same results. This is the replacement spark received from DJI after a fly away, the first one did not do this.



UPDATE: IMU calibration fixed the issue.

> the perpellas
ಠ_ಠ

I don't know about others, but that didn't seem abnormal to me.  My flights have behaved very similar.  Depending on winds and the surface over which the drone is flying, my experience has been the same.  I didn't think of this as unusual or abnormal, but the craft making necessary adjustments.

It looked fairly "tight".

Charles Adams Posted at 2017-8-7 14:04
I don't know about others, but that didn't seem abnormal to me.  My flights have behaved very similar.  Depending on winds and the surface over which the drone is flying, my experience has been the same.  I didn't think of this as unusual or abnormal, but the craft making necessary adjustments.

It looked fairly "tight".

I would agree, that is fairly consistent with what I witnessed on mine. No problems with stability while flying.

You got to be kidding, right?  Looking at the clip the drone doesn't move more than 18 inches and you're complaining???

PascalG Posted at 2017-8-7 15:42
You got to be kidding, right?  Looking at the clip the drone doesn't move more than 18 inches and you're complaining???

My first drone did not behave this way. This is the reason I asked.

Plus u are very close to the ground....ground turbulence get it up about 5 feet

Just my 2-cents, but I think this has something to do with the Spark's altitude detection. In the manual, there's a line about how visual positioning systems on page 12 (used when the drone is under ~2-4m) being affected when flying over monochrome surfaces (l.e. the black asphalt in your video). I had a very similar issue to this including the drone descending on forward stick and ascending on backwards stick in the parking lot where I live (also solid black asphalt). When I repeated the same maneuvers on grass (or any other "textured" surface), i have not seen the same behavior again and the Spark seems to hold is position better even at low altitudes. Above that height threshold, the system uses its barometer and (so far) seems to be generally more stable in all axis than near-ground. Also, as to $gambino$'s comment above, I agree that wind disturbances are definitely more noticeable visually since you're so close to the ground. 100ft up, even movements like those would be difficult to spot.

Give it a shot if you get the chance and see if the behavior is repeated over a "better" surface.

XadronsHorizon Posted at 2017-8-7 16:33
Just my 2-cents, but I think this has something to do with the Spark's altitude detection. In the manual, there's a line about how visual positioning systems on page 12 (used when the drone is under ~2-4m) being affected when flying over monochrome surfaces (l.e. the black asphalt in your video). I had a very similar issue to this including the drone descending on forward stick and ascending on backwards stick in the parking lot where I live (also solid black asphalt). When I repeated the same maneuvers on grass (or any other "textured" surface), i have not seen the same behavior again and the Spark seems to hold is position better even at low altitudes. Above that height threshold, the system uses its barometer and (so far) seems to be generally more stable in all axis than near-ground. Also, as to $gambino$'s comment above, I agree that wind disturbances are definitely more noticeable visually since you're so close to the ground. 100ft up, even movements like those would be difficult to spot.

Give it a shot if you get the chance and see if the behavior is repeated over a "better" surface.

Thanks, ill give that a try and see. i have noticed this when its high up also and it worries me like i said the first one did not act this way and wanted feedback to see what you guys saw.

if any issue, its IMU, but i doubt there is an issue.

mine i would say is about that stable maybe slightly better, but you are over pavement with patterns that are very much repeating,  which has a lot of reflected sun as well blasting back into the Visual Positioning System, so im guessing its more to do with that, it's struggling to lock in a distinct enough pattern on the pavement with the VPS sensors causing it to flutter around a bit


if it was me, i wouldnt be too worried unless struggling further up like 50 feet up..no wind and still fluttering then i would calibrate IMU

Try to calibrate the IMU, then fly the drone higher, see if it is the same.

Just piling on - this looks perfectly normal to me. Your prior flights might have been with less wind or less ground effect.

Frangu Posted at 2017-8-7 15:52
My first drone did not behave this way. This is the reason I asked.

How much accuracy do you expect from a $500 drone? It seems to be remaining within a couple of its own length. Pretty impressive to me and more than enough for smooth footage.

mmatos Posted at 2017-8-8 04:26
Just piling on - this looks perfectly normal to me. Your prior flights might have been with less wind or less ground effect.

there was no wind this day, i expected it to fly like the first spark i had. i was just concern.

PascalG Posted at 2017-8-8 04:52
How much accuracy do you expect from a $500 drone? It seems to be remaining within a couple of its own length. Pretty impressive to me and more than enough for smooth footage.

it does not worry you seeing this 100 feet off the ground and the drone moving this way with minimum wind? i would try doing the IMU calibration and see if that helps.

That just looks like VPS not locking with dark ground.

Frangu Posted at 2017-8-8 05:11
it does not worry you seeing this 100 feet off the ground and the drone moving this way with minimum wind? i would try doing the IMU calibration and see if that helps.

no, i wouldn't be too concerned that wobble is tiny. looks like VPS is struggling to make a lock, probably blinded by the sun reflecting off the pavement and pavement offering no distinct pattern to lock onto one position.     move it to a different location like somewhere with pattern for down firing sensors to lock onto i usually launch my spark if im near pavement, off sidewalk right on the lines that usually divide sidewalk "blocks"  this gives the VPS something to lockon at take off that isn't like the rest and had depth.  

if it wobbles everywhere then do an IMU calibration, but make sure calibration is done on something perfectly level and move slow.

i have done the IMU calibration and this fixed the issue.

Maybe I should do IMU calibration.  The flight looked pretty good to me.  Maybe I should expect more.

What's up on the drone?

Lian82 Posted at 2017-8-8 16:46
What's up on the drone?

that's the tile.

How to upload a video here please

Jerry Shu Posted at 2018-8-5 09:11
How to upload a video here please

You can't!
Upload to Youtube, vimeo, ...., and insert the link in your topic/post

I had THE SAME ISSUE after I had a HARD LANDING..... I replaced my props too and had no change, but then I remembered the Spark can be RECALIBRATED, and should be from time to time, so I decided to go ahead and give that a try before sending it in for repair.  No I don’t know if it makes a difference but they instruct you to take the propellers off before recalibrating so I went ahead and removed props and RECALIBRATED BOTH IMU AND COMPASS..... as soon as it finished, I replaced the props and fired her up, HOT DAMN, WOULDN'T YOU KNOW THE THING HOVERS FLAWLESSLY NOW!!  If that doesn’t work for you then I think you might have a more serious issue than you think