Was flying out over the sea today to have a closer look at some unusual boats.
All going well, in sports mode, 17 satellites, full signal.
Was about 100mtr from home on way back when screen lit up with several " compass error" " gps errors".
Still showing 17 satellites but now no signal with them.
Spark does some uncontrollable man overs before recovering and control returned to land.
Any one any ideas as to what would cause this ??

Radar interference...radars are very bad for the Spark. What kind of unusual boats were there? Military?
Do you have any pictures? :-)

Arnor Posted at 2018-3-21 12:37
Radar interference...radars are very bad for the Spark. What kind of unusual boats were there? Military?
Do you have any pictures? :-)

No not military boats it was a cable laying ship for a offshore wind farm being built.
All was ok at the closest to the ship.
Problem occurred almost back at home point which would be approx 2km from the boat

Did you use RTH on way back?

The Spark can lose signal pretty easily... be careful when flying at far distances or over other vehicles, such as boats.

Hi, I would recommend you fly in a recommended place for further check. Please do not fly too close to water which may have a bad effect on VPS. Thanks for your support.

RjMllr Posted at 2018-3-21 15:38
Did you use RTH on way back?

Only near the end but no gps by that point so no use

DJI Elektra Posted at 2018-3-21 20:23
Hi, I would recommend you fly in a recommended place for further check. Please do not fly too close to water which may have a bad effect on VPS. Thanks for your support.

Electra this is an beach with no buildings or anything at all for miles. I have flown here often with no problems at all before

N8m888 Posted at 2018-3-21 23:31
Electra this is an beach with no buildings or anything at all for miles. I have flown here often with no problems at all before

How did the sea floor look?  Clear of power cables turned on for testing?  No high voltage lines?  No large electric motors with their magnets dumped in the ocean or sunk with their boat?  Somehow your compass diverged from what the GPS and IMU was saying happened.  IMU, GPS, or Compass giving the wrong info so the autopilot hands all flying chores off to the pilot.  You have aviator sunglasses and a scarf for a reason

Keep a spare drone handy when flying over volcanoes, prisons, and the ocean (places that won't return your wayward toy).

I know a lot of Wifi Signals can also interfere with signal.  What I always do before I fly is check the wifi traffic in the area with your dji go app.  If  all channels are lit up in red probably safer to not fly.
This is one of my pre flight checks before I fly

You might want to check flight log for more info.

Montfrooij Posted at 2018-3-22 00:56
You might want to check flight log for more info.

how do i go about downloading the flight log ?

N8m888 Posted at 2018-3-22 01:07
how do i go about downloading the flight log ?

On your phone is a dji folder. in that folder is another folder named dji.go.v4   flight record is in there

N8m888 Posted at 2018-3-22 01:07
how do i go about downloading the flight log ?

I have never used them, but if you are on iOS:

Plug your iPhone into MAC / PC and run iTunes.
When device appears in iTunes at top of screen, click on this.
Under 'Summary' click on 'Apps' / (Or file share on PC)
Scroll down the main page and you'll see a section called 'file sharing', in there click on the DJI GO App, to the right you will see a list appear, in that list is a folder called 'Flight Records', click on this and then click on 'save' at the bottom of the list. This then gives you the option to save a copy of this flight record folder. Once done open up folder from the saved position and you will see all your flight record files.

On Android you can just go to the DJI GO V4 folder and select the file if you are connected to your PC.

After that go to

http://www.phantomhelp.com/LogViewer/Upload/

And upload the file.

I don’t wear a tinfoil lined hat nor do I subscribe to the many internet conspiracy theories, but something just doesn’t sit right with DJI’s lack of transparency on these type of reports.

I have many multirotor and fixed wing models employing many different GPS enabled flight controllers, - Naza V1/V2, Pixhawk, APM, Wookong, A2, Vector, etc - , and flown in what some may consider harsh RF and electromagnetic environments and never had an issue. I also have a Spark which is still on the .0700 firmware and again no issues to date.

Many of these reports have many similarities. The owner was flying in a rural area away from any visible interference sources. First they are alerted to a YAW error, then a Compass error, then loss of GPS-P forcing the craft into ATTI mode, then no response from the controller and the Spark making an abrupt change in attitude and flying off. Once the user has submitted the black box data to DJI tech support more often than not a warranty claim is upheld and the user furnished with a replacement BUT there is never a reason given as to why the Spark failed in the first place.

Without the much needed transparency needed from DJI urban myths are permitted to abound resulting in many losing all faith in their investment. It is not sufficient for a DJI rep to simply copy and paste some tenuously linked paragraph from the manual as some sort of explanation, it requires an honest response and solution based upon the evidence obtained from the autopsy of the data provided.

N8m888 Posted at 2018-3-21 23:31
Electra this is an beach with no buildings or anything at all for miles. I have flown here often with no problems at all before

If you ensured that it was not caused by the environment, you can export the flight data and the black box for analysis. You can upload it to dropbox and post the link here. Thanks for your support.

If you flew your Spark in 5.8GHz, most certainly radar interference. Today, radars are so inexpensive, that even my small 10m boat does have one together with some sonars and whatnot.

eYeSkYeYe Posted at 2018-3-22 02:28
If you flew your Spark in 5.8GHz, most certainly radar interference. Today, radars are so inexpensive, that even my small 10m boat does have one together with some sonars and whatnot.

was flying in 2.4Ghz as that is best in UK for range

satelite GPS coverage can change during the day , i always check using the uav forecast app before take off

m1n1s Posted at 2018-3-22 13:30
satelite GPS coverage can change during the day , i always check using the uav forecast app before take off

Yes but I still had 17 sats showing but with no signal ?

Have you flown in that particular area before?

davidmartingraf Posted at 2018-3-22 14:12
Have you flown in that particular area before?

Yes many times with no problems..

N8m888 Posted at 2018-3-22 14:17
Yes many times with no problems..

My first reaction is it could be a software or hardware issue, likely the reason before I would assume area interference.

Radar and possibly Loran interference that would came from the boat. Those boat-ships are fully of interferes.

Jetta-The-Hut Posted at 2018-3-22 18:22
Radar and possibly Loran interference that would came from the boat. Those boat-ships are fully of interferes.

Had this happened when I was closest to the boat I would agree, but this happened when I was almost back home which was furthest from the ship at about 2km distance.

Jetta-The-Hut Posted at 2018-3-22 18:22
Radar and possibly Loran interference that would came from the boat. Those boat-ships are fully of interferes.

Had this happened when I was closest to the boat I would agree, but this happened when I was almost back home which was furthest from the ship at about 2km distance.

Jetta-The-Hut Posted at 2018-3-22 18:22
Radar and possibly Loran interference that would came from the boat. Those boat-ships are fully of interferes.

Had this happened when I was closest to the boat I would agree, but this happened when I was almost back home which was furthest from the ship at about 2km distance.

Jetta-The-Hut Posted at 2018-3-22 18:22
Radar and possibly Loran interference that would came from the boat. Those boat-ships are fully of interferes.

Had this happened when I was closest to the boat I would agree, but this happened when I was almost back home which was furthest from the ship at about 2km distance.

I wonder if solar flare affects drone? It disrupts all electronics so I guess it's good to pay attention when the sun discharges.

Just to dispel some of these urban myths I referred to earlier in this thread.

LORAN navigation works using shore based transmitters transmitting on 100KHz. Those boats/ships that still use this system are only equipped with receivers.

Marine RADAR has two primary frequency allocations, 10cm or 3GHz and 3cm or 10GHz.

Solar flares and high K Index values have been shown to effect GPS accuracy. However studies performed demonstrated that at a K index of six and higher GPS accuracy was diminished by a factor of two to four metres in Northern Latitudes.

I have performed field strength measurements of VHF and UHF systems, using a drone to carry the test equipment, where the EIRP has been in the Kilowatt range, with no detriment to the aircrafts performance.

These very similar reports of the Spark experiencing Yaw, Compass, GPS-P, and control issues, leads me to believe that there is a possibility of errors in firmware or in hardware associated with a particular production run.  Again DJI have had the opportunity to analyse the data from many identical instances but to date have not been forthcoming with a plausible explanation.

It's been pointed out in other posts that quick turns/banks in sport mode can cause the drone to spin faster than the compass which produces the dreaded 'yaw error' and subsequent atti mode.

My Spark flew next to apartment complexes, high voltage power lines and large metallic objects without anything happening to the compass or any errors at all really. I suspect that a number of these yaw error posts are pilots pushing the Spark too far in sports mode although I could be wrong.

I'm in agreement that DJI, having received a number of flight logs already, could update their standard response templates and actually give advice on what to do to avoid such situations. Seeing as three quarters of the worlds consumer drones are produced by them with no competition in sight (or swiftly shut down) I highly doubt they'll do that.

Does anyone thing think a software update for the spark is overdue? It could fix some ATTI mode switches in mid flight