Recorded flight data vs real time data

[Scottbg1]

Hi guys,

New Mavic owner, new forum member and first time post.
First let me say I have been in RC for over 10 years. Most of that time was flying helis.
I have had quad experience for over a year. My first quad was a Yuneec Q500 with no crashes or incidences.
I have had my Mavic for a little over 2 weeks and have logged 10 flights.
I have had zero issues until today.

I took it up this morning to test some camera settings I had been playing around with.
Once everything was booted up I instantly got a warning about mag interference.
I have flown this particular spot for most of my flights and this is the first time I have received such a warning.
I thought maybe it was a high KP index as it is a moderately clear day here today so I just decided to leave the craft sitting on a platform and test the camera from the ground and fly another day.
After about 2 minutes the warning went away and the GO app gave me a green light to fly.
Just to be safe I shut down and restarted the craft and the RC and I got a green light out of the gate.

I took it up just a few feet and just hovered while I monkeyed around with the camera.
I noticed the that gimbal started to turn in a downward position by itself and it had a very bad yaw drift to the right. It has ALWAYS been rock solid on any other flight.
I allowed it to drift almost 180 degrees then I gave a corrective input to put the heading back to what it was at takeoff and I brought the camera back up to to level. The camera remained level and the drift stopped and remained stable for the rest of the flight, which was only a couple of minutes.
I didn't fly away from the original takeoff spot because I was a little spooked.
I finished my camera test and I shut it down.

I logged into HealthyDrones to see if it gave any indication of the what might have happened.
The GPS data showed a strong (green) connection the entire time, but the signal strength score was zero.
I looked back on every flight that I have logged since owning the Mavic and to my dismay I discovered that every single flight showed a very poor signal strength. Most were zero with the highest score being 24.8.

I would never fly unless the app gave me a green light to do so. The quad that I owned before I never flew with less than 12 SATs connected as thats the only indication of any kind of connection that I had with it. I've never noticed less that 14 SATs connected on the Mavic. I know SAT connection has nothing to do with RC signal, I am just throwing that in there.
My point being is that according to HealthyDrones I shouldn't have been flying, yet the GO app gave me a pass.

Am I misunderstanding the HealthyDrones data?
Do I have a problem with the Mavic?

I love this machine and I don't want to risk it or possible injury to anyone by flying stupid.
Please help.

I checked to KP index here today and it is zero.
There is a cell tower nearby where I flew today, but the other flights that showed a bad result in the past as well, I was in a rural area far away from any possible interference.
I am using the DJI GO 4 app.
Samsung Note 5 fully updated.
It's worth mentioning that I get the same green lights on Litchi as well.

Thanks in advance.

 

[Scottbg1]

Anybody?

 

[BudWalker]

Hi guys,

New Mavic owner, new forum member and first time post.
First let me say I have been in RC for over 10 years. Most of that time was flying helis.
I have had quad experience for over a year. My first quad was a Yuneec Q500 with no crashes or incidences.
I have had my Mavic for a little over 2 weeks and have logged 10 flights.
I have had zero issues until today.

I took it up this morning to test some camera settings I had been playing around with.
Once everything was booted up I instantly got a warning about mag interference.
I have flown this particular spot for most of my flights and this is the first time I have received such a warning.
I thought maybe it was a high KP index as it is a moderately clear day here today so I just decided to leave the craft sitting on a platform and test the camera from the ground and fly another day.
After about 2 minutes the warning went away and the GO app gave me a green light to fly.
Just to be safe I shut down and restarted the craft and the RC and I got a green light out of the gate.

I took it up just a few feet and just hovered while I monkeyed around with the camera.
I noticed the that gimbal started to turn in a downward position by itself and it had a very bad yaw drift to the right. It has ALWAYS been rock solid on any other flight.
I allowed it to drift almost 180 degrees then I gave a corrective input to put the heading back to what it was at takeoff and I brought the camera back up to to level. The camera remained level and the drift stopped and remained stable for the rest of the flight, which was only a couple of minutes.
I didn't fly away from the original takeoff spot because I was a little spooked.
I finished my camera test and I shut it down.

I logged into HealthyDrones to see if it gave any indication of the what might have happened.
The GPS data showed a strong (green) connection the entire time, but the signal strength score was zero.
I looked back on every flight that I have logged since owning the Mavic and to my dismay I discovered that every single flight showed a very poor signal strength. Most were zero with the highest score being 24.8.

I would never fly unless the app gave me a green light to do so. The quad that I owned before I never flew with less than 12 SATs connected as thats the only indication of any kind of connection that I had with it. I've never noticed less that 14 SATs connected on the Mavic. I know SAT connection has nothing to do with RC signal, I am just throwing that in there.
My point being is that according to HealthyDrones I shouldn't have been flying, yet the GO app gave me a pass.

Am I misunderstanding the HealthyDrones data?
Do I have a problem with the Mavic?

I love this machine and I don't want to risk it or possible injury to anyone by flying stupid.
Please help.

I checked to KP index here today and it is zero.
There is a cell tower nearby where I flew today, but the other flights that showed a bad result in the past as well, I was in a rural area far away from any possible interference.
I am using the DJI GO 4 app.
Samsung Note 5 fully updated.
It's worth mentioning that I get the same green lights on Litchi as well.

Thanks in advance.

Is the launch spot on concrete? Possibly with rebar in it? To look closer at this incident we'll need the .DAT from the Mavic (not the the .dat on the tablet). To see how to retrieve that .DAT look here. It'll be large so you'll need to Dropbox it and post the link here.

 

[Scottbg1]

Hi Bud,
Thanks so much for responding.

No, the spot that the problem occurred in is an open field.

I will get you a dropbox link for the DAT file asap.

 

[Scottbg1]

Sorry for taking so long.
My job gets in the way of my play time, sometimes.
Here's the link to the DAT file....

Dropbox - DJI_ASSISTANT_EXPORT_FILE_2017-01-04_11-04-58.DAT

 

[BudWalker]

There were two flights in the .DAT. The first one, FLY034 was actually a flight. The second flight wasn't actually a flight.

In summary, FLY034 was launched from a geomagnetically distorted site which caused the AC to be confused about it's heading.

FLY034 was launched from a geomagnetically distorted site. Almost looks like next to some metal transport containers?


When this happens the compass magnetometers have incorrect data and the heeding computed from the magnetometrs is incorrect. That incorrect heading value is then used by the Flight Controller as the initial heading. As the AC ascends it leaves the geomagnetic distortion, the compass becomes correct. BUT, the FC is now computing it's heading mostly from the IMU which is still incorrect.


As for the uncommanded yawing I'll speculate that the AC responded by yawing itself until the actual heading (according to the magnetometers) became the heading the FC wanted to see. If this is really what happened I'm quite impressed. The P3 would usually detect a problem, switch to ATTI mode, and hope things didn't get nasty. Here, the Mavic seems to be taking corrective action.

 

[Scottbg1]

Actually the map you are viewing is outdated.
Those containers were moved almost a year ago.
It's now an empty field.
However, there were some semi flatbed trailers close by.
But as I mentioned before, I've flown there before with no issues until that day.
And I apologize about both files.
I couldn't remember which file was which, and they both had a large file size so I chose them together.
I didn't know it would combine the files when downloading.

Any thoughts on my other discovery that every single flight I've logged has shown a poor signal, despite the rc showing a green light to fly?
I've seen other people's flight data and they have shown green with a signal score of 100. I've never had over 25 with the craft right in front of me.
I did read a suggestion that I might try re-binding the controller to the craft too see if that helps.

Again, thank you so much to the attention you've given this.

Sent from my SM-N920V using Tapatalk

 

[BudWalker]

Actually the map you are viewing is outdated.
Those containers were moved almost a year ago.
It's now an empty field.
However, there were some semi flatbed trailers close by.
But as I mentioned before, I've flown there before with no issues until that day.
And I apologize about both files.
I couldn't remember which file was which, and they both had a large file size so I chose them together.
I didn't know it would combine the files when downloading.

Any thoughts on my other discovery that every single flight I've logged has shown a poor signal, despite the rc showing a green light to fly?
I've seen other people's flight data and they have shown green with a signal score of 100. I've never had over 25 with the craft right in front of me.
I did read a suggestion that I might try re-binding the controller to the craft too see if that helps.

Again, thank you so much to the attention you've given this.

Sent from my SM-N920V using Tapatalk

It all depends on proximity and mass of the material causing the distortion. A difference of a few inches could make the difference.

No problem with the files, happens all the time. Pilots are still get accustomed to using DJI Assistant 2.

Can't help with the signal strength. DatCon/CsvView isn't smart enough to have that data. There have been several other pilots with RC disconnection problems. Some have claimed that a firmware update fixed the problem.