Doing a quick loop flight in an area previously miles from NFZones, at about 2:40 in this video (YouTube time) just past the top of the loop, 16 satellites and strong signal suddenly goes to ATTI mode, and I am having to fly wild. This is wilderness so also no idea why magnetic interference is also present?!

Not the most confidence building state if so... after firmware update .800 and all calibrations

What was the black item that appeared on the top of the screen ? Noticed 16 sats but it dropped to weak strength, could have been cloud interference ? It looked quite misty....TBH flying in "temple of doom" is asking for trouble lol

You almost went through Magnetic field not to sure what the dark cloud was hanging over the top of your screen, although you were reading 16 sats it looks like you only were showing 2/3 bars on the graph which is your true reading.
Lucky it didn't last to long.

hallmark007 Posted at 2017-6-15 14:46
You almost went through Magnetic field not to sure what the dark cloud was hanging over the top of your screen, although you were reading 16 sats it looks like you only were showing 2/3 bars on the graph which is your true reading.
Lucky it didn't last to long.

The dark thing on the top was the top of a gimbal guard I have, a bit like the transparent one but black with a hole where the lens sees through - so when the Mavic halts quickly the gimbal tilts up and you see the top of it. But have flown here many times, it is wilderness and never had a magnetic field warning - must be some weird software glitch?

RCNJ Posted at 2017-6-15 14:13
What was the black item that appeared on the top of the screen ? Noticed 16 sats but it dropped to weak strength, could have been cloud interference ? It looked quite misty....TBH flying in "temple of doom" is asking for trouble lol

Black item was the top of a plastic gimbal surround thing... the sun was actually shining all the time hence the start of the flight completely full strength 16 sats - there was some mist 2km away but none here. I am troubled that it goes from no interference, full sats, clear sky - perfect conditions yet it just drops GPS?!  

What was the solar weather looking like that day/time? Ah, check this out: http://www.sws.bom.gov.au/Geophysical/2/1
Geomagnetic Warning

(last updated 15 Jun 2017 23:45 UT)

SUBJ: SWS GEOMAGNETIC DISTURBANCE WARNING 17/31 ISSUED AT 2311UT/14 JUNE 2017 BY THE AUSTRALIAN SPACE FORECAST CENTRE.  In the second half of the first UT day (15 June) the solar wind  is expected to be influenced by a High Speed Solar Wind Stream.  On the second UT day (16 June) the speed can reach 700 km/s.  The geomagnetic activity is expected to reach active levels.  Isolated minor storm periods are possible on 16 June.  INCREASED GEOMAGNETIC ACTIVITY EXPECTED  DUE TO CORONAL HOLE HIGH SPEED WIND STREAM FROM 15-16 JUNE 2017 _____________________________________________________________  GEOMAGNETIC ACTIVITY FORECAST 15 Jun:  Unsettled to Active 16 Jun:  Active

Ex Machina Posted at 2017-6-15 15:57
What was the solar weather looking like that day/time? Ah, check this out: http://www.sws.bom.gov.au/Geophysical/2/1
Geomagnetic Warning[view_image]
(last updated 15 Jun 2017 23:45 UT)

This was filmed about 7.30am on 14 June - but do Geomagnetic events seriously affect drones?! If that happens surely hundreds of airborn Dji drones will suddenly lose GPS lock and start to 'drift', multiple crashes etc? Will have to look into this more... another reason to keep the drone grounded! along with rain, mist, wind, too cold, too hot, new NFZ etc etc  

Another thought- https://www.usatoday.com/story/n ... west-june/85562276/

Last year I flew during this test and didn't notice anything. https://www.faasafety.gov/files/ ... Flight_Advisory.pdf

These notices come out all the time all over the country.

If you are all "tin foil hat" chalk it up to aliens or magic.

If you are "political" blame Trump, Hillary, or the Russians.

Truth is that no radio communications are 100%. It's just a matter of how fault tolerant the system is.

garyphayes Posted at 2017-6-15 16:10
This was filmed about 7.30am on 14 June - but do Geomagnetic events seriously affect drones?! If that happens surely hundreds of airborn Dji drones will suddenly lose GPS lock and start to 'drift', multiple crashes etc? Will have to look into this more... another reason to keep the drone grounded! along with rain, mist, wind, too cold, too hot, new NFZ etc etc

Yes, if the storms are high enough on the KP index. More info here -- they also have a nice app that lets you check this kind of thing on the fly:

https://www.uavforecast.com/#/faq

Ex Machina Posted at 2017-6-15 17:24
Yes, if the storms are high enough on the KP index. More info here -- they also have a nice app that lets you check this kind of thing on the fly:

https://www.uavforecast.com/#/faq

Thanks for that. Are there examples where this has actually brought drones down or caused crashes? Is it well known in the drone fraternity, should beginners be aware of it too?

Glenn Goodlett Posted at 2017-6-15 16:31
Another thought- https://www.usatoday.com/story/news/2016/06/07/faa-military-testing-could-leave-gps-unreliable-pilots-across-west-june/85562276/

Last year I flew during this test and didn't notice anything. https://www.faasafety.gov/files/notices/2016/Jun/CHLK_16-08_GPS_Flight_Advisory.pdf

Thanks for that - have retired my tin foil hat. I suppose it is more a question that have been flying the Mavic since early Jan and not had one issue. No drop out, rock solid hover, never seen any NFZ or magentic interference warnings until this firmware update. Have Dji got a bug where an NFZ warning triggers magenetic interference and/or GPS dropout?

garyphayes Posted at 2017-6-15 17:26
Thanks for that. Are there examples where this has actually brought drones down or caused crashes? Is it well known in the drone fraternity, should beginners be aware of it too?

I seems to be well known in the pilot and drone community -- I learned about it here. How often flyaway and other problems are caused by high KP index days I couldn't say.

Seems like another data point to consider if you are planning on flying long distances and/or out of LOS, or trying to diagnose odd behavior.

I think DJI basically made the drone limit itself when no GPS is present. Seems like they are doing this so drones can't be wrapped with tin foil and used to deliver hand grenades and that sort of thing. I think the NFZ message was a response to losing GPS lock. I could be wrong as well.

I have seen some random magnetic interference warnings and quick GPS drop outs on previous firmware versions, but they were pretty rare. The NFZ message is definitely new with the most recent firmware.

garyphayes Posted at 2017-6-15 17:34
Thanks for that - have retired my tin foil hat. I suppose it is more a question that have been flying the Mavic since early Jan and not had one issue. No drop out, rock solid hover, never seen any NFZ or magentic interference warnings until this firmware update. Have Dji got a bug where an NFZ warning triggers magenetic interference and/or GPS dropout?

If you've had it since early January you may have the old cabling that was more prone to generating interference that could trigger ATTI mode in situations like this. If you're comfortable with do-it-yourself solutions you may want to check out this thread. In any case if you'd like to post the DAT file from the flight I or someone else can take a look at it and may be able to give you a better idea of what happened.

Use these instructions to access the DAT file and once you've done so you'll need to put it in DropBox and post a link to it in this thead.

Glenn Goodlett Posted at 2017-6-15 17:52
I think DJI basically made the drone limit itself when no GPS is present. Seems like they are doing this so drones can't be wrapped with tin foil and used to deliver hand grenades and that sort of thing. I think the NFZ message was a response to losing GPS lock. I could be wrong as well.

I have seen some random magnetic interference warnings and quick GPS drop outs on previous firmware versions, but they were pretty rare. The NFZ message is definitely new with the most recent firmware.

But in ATTI mode you can still fly unhindered, just need to be much more responsive as a pilot and be aware that wind can push it anywhere?

Ex Machina Posted at 2017-6-15 17:48
I seems to be well known in the pilot and drone community -- I learned about it here. How often flyaway and other problems are caused by high KP index days I couldn't say.

Seems like another data point to consider if you are planning on flying long distances and/or out of LOS, or trying to diagnose odd behavior.

I also didn't realise in ATTI mode that RTH becomes inoperative - scary stuff

DroneFlying Posted at 2017-6-15 18:05
If you've had it since early January you may have the old cabling that was more prone to generating interference that could trigger ATTI mode in situations like this. If you're comfortable with do-it-yourself solutions you may want to check out this thread. In any case if you'd like to post the DAT file from the flight I or someone else can take a look at it and may be able to give you a better idea of what happened.

Use these instructions to access the DAT file and once you've done so you'll need to put it in DropBox and post a link to it in this thead.

Thanks DF - will give the dropbox links to the 3 DAT files in a moment

If I start fiddling with rev compass, doesn't that void any warranty, given the Mavic is only 6 months old?

DAT FILES - NOTE WRONG ONES. PROPER ONE BELOW

There were three generated from this flight at 7.39am on 14 Jun (think 258 is the main one at 8MB)
256 https://www.dropbox.com/s/yol8tmi2pshv2kk/FLY256.DAT?dl=0
257 https://www.dropbox.com/s/qxuy6nh9czcz6ca/FLY257.DAT?dl=0
258 https://www.dropbox.com/s/rcu4yetp26ksly4/FLY258.DAT?dl=0

garyphayes Posted at 2017-6-15 18:11
I also didn't realise in ATTI mode that RTH becomes inoperative - scary stuff

Makes sense right? If the Mavic doesn't know where it is, it doesn't know where home is either.
I do agree though that high kp will likely affect all transmission, not just GPS signals, but you may also experience signal loss from the Mavic. So be aware, and be ready to take over manually.

PHL Posted at 2017-6-15 18:54
Makes sense right? If the Mavic doesn't know where it is, it doesn't know where home is either.
I do agree though that high kp will likely affect all transmission, not just GPS signals, but you may also experience signal loss from the Mavic. So be aware, and be ready to take over manually.

Thanks. No more filming at 1.5km away then - if it can just lose it. Or at least be high enough so if it does you can recover. Suddenly after 6 months of confident flying with no odd dropouts I am now like, best fly 20m above everything, in straight lines and don't go more than 200m  away Hello drone stress...

garyphayes Posted at 2017-6-15 18:35
DAT FILES

There were three generated from this flight at 7.39am on 14 Jun (think 258 is the main one at 8MB)

Could you check again? None of those appears to be the correct one and in fact the only one that's viewable is 258, which appears from powering on the aircraft for about a minute indoors. The correct file -- that is, the one from the flight you mentioned -- is going to be quite a bit larger than the ones you already uploaded.

DroneFlying Posted at 2017-6-16 02:33
Could you check again? None of those appears to be the correct one and in fact the only one that's viewable is 258, which appears from powering on the aircraft for about a minute indoors. The correct file -- that is, the one from the flight you mentioned -- is going to be quite a bit larger than the ones you already uploaded.

Ah found them - uploading now. Thanks for spotting!

CORRECT DAT FILE/S 250MB

https://www.dropbox.com/s/spx4j0 ... 6_21-32-57.DAT?dl=0

garyphayes Posted at 2017-6-16 03:46
CORRECT DAT FILE/S 250MB

https://www.dropbox.com/s/spx4j0lpxyz7sxj/DJI_ASSISTANT_EXPORT_FILE_2017-06-16_21-32-57.DAT?dl=0

Yep, that's the right one. I've looked at it and I suspect that it was caused by the rear-compass-noise-induced-ATTI-mode problem discussed in the thread I linked to earlier. As the picture below shows, the rear compass (Mag:Z:Back) is experiencing a bit more "noise" than the front (Mag:Z:Front), and your aircraft's build date of November 3, 2016 indicates that it wouldn't have included the cabling change that DJI made in newer Mavics to alleviate the problem. I'd like to see BudWalker weigh in on this, though, because he has more experience recognizing this problem and might have a different diagnosis.

Assuming my analysis is correct, your options for addressing the problem are to send it in to DJI for a warranty repair (more likely a replacement), to perform the cabling twist yourself using the instructions in that other thread, or to pay a repair shop to do it for you.

P.S. Even though the NFZ error showed up first, I suspect that Glenn is correct in that it was triggered by the ATTI problem and not the other way around.

garyphayes Posted at 2017-6-15 19:00
Thanks. No more filming at 1.5km away then - if it can just lose it. Or at least be high enough so if it does you can recover. Suddenly after 6 months of confident flying with no odd dropouts I am now like, best fly 20m above everything, in straight lines and don't go more than 200m  away  Hello drone stress...

You're not alone, I'm feeling "Drone Stress" too.

For those who were surprised that loss of GPS would cause loss of RTH, please remember, the drone is dumb.  It has no idea where it is.  The GPS is the only thing in it that knows where it is.  If the GPS fails, anything that relies on GPS will fail also.  RTH fails, position hold fails, waypoints fail, etc.

Now, I had a mind to anyway, but I guess I had better think harder about doing the wire twist mod also.  Mine was built in November.  

DroneFlying Posted at 2017-6-16 05:47
Yep, that's the right one. I've looked at it and I suspect that it was caused by the rear-compass-noise-induced-ATTI-mode problem discussed in the thread I linked to earlier. As the picture below shows, the rear compass (Mag:Z:Back) is experiencing a bit more "noise" than the front (Mag:Z:Front), and your aircraft's build date of November 3, 2016 indicates that it wouldn't have included the cabling change that DJI made in newer Mavics to alleviate the problem. I'd like to see BudWalker weigh in on this, though, because he has more experience recognizing this problem and might have a different diagnosis.

Assuming my analysis is correct, your options for addressing the problem are to send it in to DJI for a warranty repair (more likely a replacement), to perform the cabling twist yourself using the instructions in that other thread, or to pay a repair shop to do it for you.

If this were the problem, wouldn't it have presented before now?

I would think so if he flew exactly the same from the same location.

Ex Machina Posted at 2017-6-16 09:04
If this were the problem, wouldn't it have presented before now?

But no real issues before the .800 firmware update... is the new software to blame in combination?

AG0N-Gary Posted at 2017-6-16 12:45
I would think so if he flew exactly the same from the same location.

Yes on different firmware...

AG0N-Gary Posted at 2017-6-16 07:10
For those who were surprised that loss of GPS would cause loss of RTH, please remember, the drone is dumb.  It has no idea where it is.  The GPS is the only thing in it that knows where it is.  If the GPS fails, anything that relies on GPS will fail also.  RTH fails, position hold fails, waypoints fail, etc.

Now, I had a mind to anyway, but I guess I had better think harder about doing the wire twist mod also.  Mine was built in November.

Yes pretty disastrous scenario if in a closed in environment too - it is the drift factor in wind that scares me most and relying on ones reaction to realise that you have to take over and fly it fully manual

DroneFlying Posted at 2017-6-16 05:47
Yep, that's the right one. I've looked at it and I suspect that it was caused by the rear-compass-noise-induced-ATTI-mode problem discussed in the thread I linked to earlier. As the picture below shows, the rear compass (Mag:Z:Back) is experiencing a bit more "noise" than the front (Mag:Z:Front), and your aircraft's build date of November 3, 2016 indicates that it wouldn't have included the cabling change that DJI made in newer Mavics to alleviate the problem. I'd like to see BudWalker weigh in on this, though, because he has more experience recognizing this problem and might have a different diagnosis.

Assuming my analysis is correct, your options for addressing the problem are to send it in to DJI for a warranty repair (more likely a replacement), to perform the cabling twist yourself using the instructions in that other thread, or to pay a repair shop to do it for you.

OK. When I get my new P4P back from Dji which had compass 1 issues and dodgy remote, I will send the Mavic back for warranty if this is confirmed. I guess the fault could have surfaced with the .800 firmware update given it has not done this before? And thanks for looking into it!!

jamesw Posted at 2017-6-16 06:27
You're not alone, I'm feeling "Drone Stress" too.

I was relatively drone stress free for the first 5 months but one would think $2000 (with extras) pieces of equipment would be highly reliable... sadly not so

garyphayes Posted at 2017-6-16 14:54
I was relatively drone stress free for the first 5 months but one would think $2000 (with extras) pieces of equipment would be highly reliable... sadly not so

Well hopefully you get sorted , actually when your drone is out there and if it happens to go down all the extras and your RC won't go down with it, so maybe a loss of $750 unless its malfunction, then it's a warranty.

hallmark007 Posted at 2017-6-16 15:03
Well hopefully you get sorted , actually when your drone is out there and if it happens to go down all the extras and your RC won't go down with it, so maybe a loss of $750 unless its malfunction, then it's a warranty.

True. I was in AUD $ - but yes assuming you can get the body back for warranty. But supposing not, then the 64GB sd card, polar pro filter, battery etc: add up too...

garyphayes Posted at 2017-6-16 14:54
I was relatively drone stress free for the first 5 months but one would think $2000 (with extras) pieces of equipment would be highly reliable... sadly not so

I guess you were not around when the DJI Spreading Wings, A2 guided drones would randomly do a flip of death with your $10,000 drone. That was stress.

Glenn Goodlett Posted at 2017-6-16 17:54
I guess you were not around when the DJI Spreading Wings, A2 guided drones would randomly do a flip of death with your $10,000 drone. That was stress.

Yes all a matter of perspective   Also one persons $10k is anothers $1k  is anothers $100

garyphayes Posted at 2017-6-16 14:48
But no real issues before the .800 firmware update... is the new software to blame in combination?

Seems unlikely, the wiring noise interference source is in hardware, if anything, I'd expect firmware released after the issue was discovered to be more resistant. Anyway, no rash in unexpected ATTI switching post .0800 reported here as far as I can tell.

Might also be the newer firmware is more strict on what constitues a good GPS signal, and what doesn't. Before there might have been aime fudging, but there may have been other problems with that strategy in some circumstances, so the parameters might have been tightened.
I've certainly seen more GPS loss and low signal warnings then before, but usually for a second or so, although where I am is close to a radio waves research lab, so they may have also been doing some testing.

Ex Machina Posted at 2017-6-15 17:24
Yes, if the storms are high enough on the KP index. More info here -- they also have a nice app that lets you check this kind of thing on the fly:

https://www.uavforecast.com/#/faq

i thought this wasnt relevant anymore because of glonass