Please select Into the mobile phone version | Continue to access the computer ver.
Dji Spark Compass error in flight , exit-gps
123Next >
11400 96 2017-9-12
Uploading and Loding Picture ...(0/1)
o(^-^)o
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

Guys, just had a similar problem today. I've had about 38 flights and Today flying my spark in almost perfect conditions, had 14 satellites locked before I took off and recorded my home point, then just a few seconds into the flight at 33m high I lost all GPS signal and the spark flew away 110m and went into Atti Mode. Luckily I was able to take control once GPS returned Any suggestions?  See video of flight log. Ps. I love my spark but this has me a little worried.
2017-10-1
Use props
djiuser_xLE6O03
Second Officer
Flight distance : 1198346 ft
Romania
Offline

Can you post your dat file ?
Also i have 20 flight without troubles since the error but I am worried too
2017-10-1
Use props
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

Hi Mate/Community,

Please see my flight log uploaded to the to the online DJI Flight Log Viewer (http://www.phantomhelp.com/LogViewer/U21QRV9CPDVGQKU3BAIV/). As you can see about 17s into the flight at around 100ft, I get:

1.Yaw Error, followed by;
2. Compass Error. Exit P-GPS Mode; Yaw Error, followed by;
3. In Flight,working,IMU,encountersheading exception,please switch to atti mode if craft behave abnormally; Yaw Error.

Fortunately, I am able to recover about 39 secs into the flight but for the 22 secs it drifted 320ft.

All of this happened whilst having 14 satellites supposedly. I was flying in park land, clear of any interruptions (I think). Any ideas?





2017-10-2
Use props
djiuser_xLE6O03
Second Officer
Flight distance : 1198346 ft
Romania
Offline

Please post dat file from spark
2017-10-2
Use props
Charles Adams
First Officer
Flight distance : 3821312 ft
  • >>>
United States
Offline

bfreedman Posted at 2017-10-2 03:17
Hi Mate/Community,

Please see my flight log uploaded to the to the online DJI Flight Log Viewer (http://www.phantomhelp.com/LogViewer/U21QRV9CPDVGQKU3BAIV/). As you can see about 17s into the flight at around 100ft, I get:

On short review, I can tell you what happened, though I cannot tell you why it happened.

As Hallmark explains (I always try to credit his knowledge), when a craft is getting conflicting data from the compass and from GPS, it will abandon one over the other.  It will always drop GPS in favor of the compass, because the craft cannot be flown with GPS alone, the compass is required.  Once in ATTI mode it becomes challenging to control the craft, because the environment has greater impact on the flight.  It appears that there were some winds that day (that's just a guess from the speeds attained), which the pilot would have to contend with.  That's the "what" happened.

Why did it happen?  Why were there compass issues early on?  The only thing that jumps out at me from the map is the light pole which appears to be (guessing 50-100 feet) in proximity to your launch location.  But that's the only man-made structure I see that is possibly close enough to have caused interference.  In my mind the question isn't whether or not it could cause interference (yes it absolutely could), the question is were you close enough to have had it influence your flight.  I personally would have probably launched a bit farther away (I try to be 500 feet away from power lines and other powered structures), but I can't claim that you were too close.  Were the lights on at the time you were flying?  If they were (if power was flowing), then I would be more comfortable in drawing a conclusion specific to that object.  Big lights can generate lots of interference when on.  I would go from "maybe" to "most probably".  If it were off, then I'd be more in doubt.

One thing you can do to test this is to launch in the same area, but farther away and see if you have a repeat of the issues.  If you are bold you could try from the exact same location, but I wouldn't tempt fate.
2017-10-2
Use props
djiuser_Kf4iPA3
First Officer
Flight distance : 510968 ft
Israel
Offline

So it has a compass error, and the result is to drop all other data (much more reliable data, mind you) and rely on the erroneous compass...
I'm shocked that this makes sense to so many people.

I understand that's what happens now, I don't dispute that. But I dispute the logic of that. I disagree that the aircraft cannot orient itself in an emergency based on GPS alone, at least to be able to head in the direction of home and the pilot. Yes it can, although it takes some time and some zigzaging. In another thread I explained how.
2017-10-2
Use props
djiuser_xLE6O03
Second Officer
Flight distance : 1198346 ft
Romania
Offline

bfreedman Posted at 2017-10-2 03:17
Hi Mate/Community,

Please see my flight log uploaded to the to the online DJI Flight Log Viewer (http://www.phantomhelp.com/LogViewer/U21QRV9CPDVGQKU3BAIV/). As you can see about 17s into the flight at around 100ft, I get:

22 secs also in my case
2017-10-2
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline

bfreedman Posted at 2017-10-2 03:17
Hi Mate/Community,

Please see my flight log uploaded to the to the online DJI Flight Log Viewer (http://www.phantomhelp.com/LogViewer/U21QRV9CPDVGQKU3BAIV/). As you can see about 17s into the flight at around 100ft, I get:

Can you provide a Dropbox or GoogleDrive link to the .DAT file for this flight? To see how to retrieve the .DAT look here

http://www.datfile.net/DatCon/retrieveV3Dat.html
2017-10-2
Use props
jcwconsulting
lvl.4
Flight distance : 944236 ft
United States
Offline

Saint74 Posted at 2017-9-15 00:04
Nothing  All was ok. 19 satellites. Then all went wrong. Just for a few second.

yup - i know those errors all too well...same set that i got on 9/4 when my Spark originally freaked before the first warranty repair claim during which they replaced the GPS receiver
2017-10-2
Use props
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

Hi Guys,

Thanks for your input so far.

  • @BudWalker and djiuser_xLE6O03, I will post my .dat file but unfortunately this will have to wait until Tommorow.
  • @Charles Adams, thanks for your insight. I must admit, I too suspected that it may have been the light pole near from where I took off; I guess what I am curious on is why on the return flight home (similar path to where I took off), there was no interference? As a side note, the lights were off.

Ultimately however, what I am more interested in is the best practice that can be applied in situations like these. I accept there is always going to be a degree  of risk that technology or enviromental factos that may result in such incidents, but what I think the important thing to take away is the best practices that can be applied when they do.

In my case, I noticed that it went into ATTI mode, I did the following things (this includes my questions and commentary), hopefully we can take something away....

  • I immediately climbed in altitude (within my local guidelines), to reduce the likelihood of my AC drifting into any low objects and crashing down onto anything. Is this recommended?
  • I was also wary not to press the RTH button (although) not sure this would have worked because the GPS was disabled, however if I did, could this have made the situation worse?
  • I also have a question surrounding handling of the AC when in ATTI mode, my understanding is that the only the X (Forward/Back) and Y (Left/Right) axis are not controlled. In my case I also had a Yaw Error. At the distance the drone was, it's quite hard to fly LOS and to see the front of the AC. To contend with the natural movements of the AC, do you think it would be better approach to fly back, zig-zagging home using the FPV display (while resisting the AC's natural movement).
  • If I didn't do anything but let the AC drift in ATTI mode (assuming it cleared all objects), would the AC not stabilise itself as it is naturally drifting and potentially flowing away from the source of interference; in this case would it not just switch into GPS mode? Obviously the risk here is that it drifts too far and although it returns to GPS mode, perhaps the transmission is lost, on that note; would it not try and RTH?
  • What is the best practise/what thresholds do you guys use to observe compass interference prior to take off?
  • Is there anything else that one can do to pro-actively recover the AC once it is in ATTI mode?
2017-10-2
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline

bfreedman Posted at 2017-10-2 21:51
Hi Guys,

Thanks for your input so far.

To answer your questions I took a closer look at the .txt. The .DAT will possibly tell us some more...
1) This does not appear to be a launch-from-a-geomagnetically-distorted-site incident. Reviewing the initial values for stick inputs, Yaw values, and direction of flight indicates that Yaw was correct. The AC appears to have been pointing in a SW direction at launch. Is this correct?
2) There wasn't a loss of satellites during the flight but there was a loss of gpsHealth in the interval 18 secs - 43 secs which resulted in the switch to ATTI. At the beginning (18 secs) the Spark was still over the launch site and had done most of the climbing for this part of the flight. This can be seen in the attach #1
3) Although it may have seemed like the Spark was not controllable during the ATTI interval it did respond appropriately to control inputs. But, it was also drifting with the wind out of the SW. To see this it's easiest if you use CsvView locally - it's hard to show with a static display. But try this. Attach #2 shows an interval where elevator and aileron are 0 and attach #3 shows the Spark drifting NE

As for your questions
2) You're right. Under no circumstances should you initiate RTH when there is a compass error. And, hope that there isn't a RC disconnect which will then cause an RTH.
3) & 4) As I stated I don't think this was a compass issue. It was a GPS issue. But, yes, if you can't determine which way it's pointing to fly it back home it's better to let it drift and hope it returns to GPS+ATTI mode. One method you can use is to give it some elevator input and then observe it's location change on the map. You can then determine which way it's pointing.
5) The single most effective way to prevent launch time flyaways is to check that the red triangle heading indicator in the map display is consistent with the actual AC orientation


bfreedman1.jpg
bfreedman2.jpg
bfreedman3.jpg
2017-10-3
Use props
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

Thank you for your insight. This is very useful. I have now been playing with CsvView, and trying to understand the State and Series signals mean. I can guess most of it, but do you know if there is a glossary of what some of the non-obvious signals mean? I observed the nonGPSCause was YAW_ERROR_LARGE and COMPASS_ERROR_LARGE; this this a product of the AC going into ATTI mode or the reverse, it went into ATTI mode because of these things? When it did I noticed a larger variance between the Yaw MagYaw, again don't know if this means anything until I familiarize myself with these signals; just trying to learn and understand as much as I can

My .dat file is now on google drive, if you want to have a look: https://drive.google.com/open?id=0B-_f1510zCO9NTFKSkVEWElrUHc
2017-10-4
Use props
Pirikitto
lvl.3
Flight distance : 183658 ft
United States
Offline

bfreedman Posted at 2017-10-4 03:34
Thank you for your insight. This is very useful. I have now been playing with CsvView, and trying to understand the State and Series signals mean. I can guess most of it, but do you know if there is a glossary of what some of the non-obvious signals mean? I observed the nonGPSCause was YAW_ERROR_LARGE and COMPASS_ERROR_LARGE; this this a product of the AC going into ATTI mode or the reverse, it went into ATTI mode because of these things? When it did I noticed a larger variance between the Yaw MagYaw, again don't know if this means anything until I familiarize myself with these signals; just trying to learn and understand as much as I can  

My .dat file is now on google drive, if you want to have a look: https://drive.google.com/open?id=0B-_f1510zCO9NTFKSkVEWElrUHc

Hi, I'm not an expert but from the chart I think your compass is not calibrated
If you see the MagYaw line is too distracted and distant from the Yaw, in to my chart the two lines are virtually identical
Yaw-MagYaw.jpg

Instead MagData is all in the south, in my DAT file is in North, I do not know what it means, we wait for some experts!

MagData1.jpg MagData2.jpg
2017-10-4
Use props
Charles Adams
First Officer
Flight distance : 3821312 ft
  • >>>
United States
Offline

bfreedman Posted at 2017-10-4 03:34
Thank you for your insight. This is very useful. I have now been playing with CsvView, and trying to understand the State and Series signals mean. I can guess most of it, but do you know if there is a glossary of what some of the non-obvious signals mean? I observed the nonGPSCause was YAW_ERROR_LARGE and COMPASS_ERROR_LARGE; this this a product of the AC going into ATTI mode or the reverse, it went into ATTI mode because of these things? When it did I noticed a larger variance between the Yaw MagYaw, again don't know if this means anything until I familiarize myself with these signals; just trying to learn and understand as much as I can  

My .dat file is now on google drive, if you want to have a look: https://drive.google.com/open?id=0B-_f1510zCO9NTFKSkVEWElrUHc

I don't know that my processes are the best, but here's what I do to reduce the risk of losing the craft.

I pay attention to any warnings offered in the software.  If I have a warning or error, I don't fly until I've resolved that warning or error.

I don't fly in extreme weather.  10mph winds max.

I only fly where I can legally retrieve the craft should it ditch.

I only fly where I have confidence that there are not any nearby man-made influences (light poles, power lines, etc).

I only fly line of sight.  In the event that the craft goes into atti mode (and it has for me) my eyes become my navigation system.

In atti mode I elevate the craft in the hopes of getting clearer gps signal (and/or leaving compass interference).  This also helps to keep LOS on the craft.  I use to have a different plan, but based on input from participants in this forum, I've changed this behavior.

I don't panic (or try not to)!  Now this is really a hard one to control.  Panic is a biological reaction to stress, and (thank you Darwin), it causes physiological changes in how your brain operates.  These changes can help you in other situations, but they don't help you fly.  It's very difficult to control this reaction, but there are ways to prepare for it.  I practice maneuvers ahead of time (I have a P3S that I can use for atti practice flights), so that in the event I face a stressful circumstance in flight, I have some level of personal training and experience to draw upon when reacting.  This helps mitigate the effects of panic.
2017-10-4
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline

bfreedman Posted at 2017-10-4 03:34
Thank you for your insight. This is very useful. I have now been playing with CsvView, and trying to understand the State and Series signals mean. I can guess most of it, but do you know if there is a glossary of what some of the non-obvious signals mean? I observed the nonGPSCause was YAW_ERROR_LARGE and COMPASS_ERROR_LARGE; this this a product of the AC going into ATTI mode or the reverse, it went into ATTI mode because of these things? When it did I noticed a larger variance between the Yaw MagYaw, again don't know if this means anything until I familiarize myself with these signals; just trying to learn and understand as much as I can  

My .dat file is now on google drive, if you want to have a look: https://drive.google.com/open?id=0B-_f1510zCO9NTFKSkVEWElrUHc

This is the weirdest, most complicated .DAT that I've ever seen. There is more than one problem so I'm going to have to go with I don't know what the problem is. But, I'll give you my observations. I'd tend to think the problems are related and stem a single underlying hardware issue. My advice is that it needs to go to DJI repair.
Can you provide the .DATs that include the previous flight and all the .DATs in between? I'd be interested to see if any of those show the kinds of problems seen in FLY058


At time -65.488 secs (time 0.0 is when the flight starts) there was a glitch, spike, that affected all of the IMU values. E.g. GyroZ had a value of 3,219,351,142,400. Never seen this before and it didn't happen after that. I don't know if this is just a one-off and can be ignored or it's an indicator that of a hardware issue.

At time 17.736 there was a YAW_ERROR_LARGE error for no apparent reason (at least not apparent to me). The gpsHealth was set to 0 followed by the switch ATTI mode. I kinda suspect that there wasn't anything wrong with the GPS and setting it to 0 might be what triggers the ATTI mode. This seemed to be how the Mavic worked a while back when there were many gpsHealth-drops-to-0 incidents. see attach 1

As Pirikitto noted there were several times where a COMPASS_ERROR_LARGE was triggered by a separation of magYaw and Yaw. (For more info on magYaw look here https://phantompilots.com/threads/momentary-control-loss-p3p.123967/#post-1265599) In fact the pattern looks very much like the compass isn't calibrated properly. See attach 2. In the region 57 to 114 degrees magYaw is greater than Yaw and in the region 0 to -114 degrees magYaw is less than Yaw. I've done experiments with my P3 where a compass calibration was done with an allen wrench attached close to the compass. Removing the allen wrench produced a P3 with a bad calibration and a Yaw/magYaw pattern similar to yours.

Have you attached an external device such as a tracker?



bfreedman4.jpg
bfreedman5.jpg
2017-10-4
Use props
Pirikitto
lvl.3
Flight distance : 183658 ft
United States
Offline

BudWalker Posted at 2017-10-4 16:18
This is the weirdest, most complicated .DAT that I've ever seen. There is more than one problem so I'm going to have to go with I don't know what the problem is. But, I'll give you my observations. I'd tend to think the problems are related and stem a single underlying hardware issue. My advice is that it needs to go to DJI repair.
Can you provide the .DATs that include the previous flight and all the .DATs in between? I'd be interested to see if any of those show the kinds of problems seen in FLY058

Thanks for explanation BudWalker

I have a question, why his MagData sphere is at Sud pole instead my is at North pole?
Thanks
2017-10-5
Use props
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

Thanks for the insight guys, a little tied up tonight but will post my .dat file Tommorow evening.
2017-10-5
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline

Pirikitto Posted at 2017-10-5 00:28
Thanks for explanation BudWalker

I have a question, why his MagData sphere is at Sud pole instead my is at North pole?

The documentation for the MagData Player can be found here
http://www.datfile.net/Doc/MagDataPlayer.pdf

In general, in the southern hemisphere the geomagnetic flux will be coming from underneath the AC; i.e. the negative Z direction.  Closer to the south magnetic pole the geomagnetic flux will be more aligned with the negative Z axis.
2017-10-5
Use props
Pirikitto
lvl.3
Flight distance : 183658 ft
United States
Offline

BudWalker Posted at 2017-10-5 03:30
The documentation for the MagData Player can be found here
http://www.datfile.net/Doc/MagDataPlayer.pdf

I get it! Thanks!!
2017-10-5
Use props
SparkeD
New
Flight distance : 71650 ft
Canada
Offline

I have experienced this ATTI mode, no GPS signal and erratic flight response twice now. Once in August when flying above a lake and just yesterday when flying neat a bike trail.  The drone seems to get internally switched to sport mode and does not respond to remote control commands.  Luckily the Spark only clipped a tree and I was able to regain control after a short period of uncontrolled flight. This is very nerve racking.
IMG_6053.PNG
2017-10-5
Use props
Pirikitto
lvl.3
Flight distance : 183658 ft
United States
Offline

SparkeD Posted at 2017-10-5 06:10
I have experienced this ATTI mode, no GPS signal and erratic flight response twice now. Once in August when flying above a lake and just yesterday when flying neat a bike trail.  The drone seems to get internally switched to sport mode and does not respond to remote control commands.  Luckily the Spark only clipped a tree and I was able to regain control after a short period of uncontrolled flight. This is very nerve racking.

Upload your DAT file!
2017-10-5
Use props
SparkeD
New
Flight distance : 71650 ft
Canada
Offline


I think I have the DAT file uploaded to dropbox. There does not appear to be a DJI Assistant for the MAC OS. Is there a way to view/investigate these files on a MAC?

DAT File
2017-10-5
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline

SparkeD Posted at 2017-10-5 13:50
I think I have the DAT file uploaded to dropbox. There does not appear to be a DJI Assistant for the MAC OS. Is there a way to view/investigate these files on a MAC?

DAT File

Not sure where that .dat came - maybe the RC? Use these instructions to retrieve the .DAT from the Spark itself.

http://www.datfile.net/DatCon/retrieveV3Dat.html

You might want to try the CsvView Mac version. Go to
http://www.datfile.net/CsvView/downloads.html
and download the CsvViewMac version. There are instructions at the bottom of that page explaining how to run the Mac version.
2017-10-5
Use props
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

bfreedman Posted at 2017-10-5 02:55
Thanks for the insight guys, a little tied up tonight but will post my .dat file Tommorow evening.

Hi Guys,





What I've observed in both these flights was that the variance between the Yaw and MagYaw values seems to be significantly smaller; especially in comparison to the problem flight, I observed looking at the graph and the GeoPlayer that at the time of my issue the AC starting spinning around in circles; in ATTI mode and when your drone is spinning, I imagine it would be extremely difficult to control and recover. Based on my limited assesment, I'd probably conclude that it was an environment issue and for whatever reason there appears to have been some very high degree of electro magnetic interference. Thoughts?

2017-10-6
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline


Thanks. But could you please retrieve the .DAT files for the flight before the incident flight which was FLY058. That may not be FLY057.DAT since the Spark will create a .DAT every time the battery is turned on even if there is no flight.  Also, if there are any .DAT files between the the previous flight and FLY058.DAT please provide those as well.

I'm interested to know if the problems existed before the incident flight.
2017-10-6
Use props
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

BudWalker Posted at 2017-10-6 05:42
Thanks. But could you please retrieve the .DAT files for the flight before the incident flight which was FLY058. That may not be FLY057.DAT since the Spark will create a .DAT every time the battery is turned on even if there is no flight.  Also, if there are any .DAT files between the the previous flight and FLY058.DAT please provide those as well.

I'm interested to know if the problems existed before the incident flight.

I can't unfortunately, to my knowledge the hard drive onboard the AC only cache's so many logs, since the time of the incident, I did turn on the device a few times trying to investigate some things, so got rid of the older records, unless I'm wrong?
2017-10-6
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline

bfreedman Posted at 2017-10-6 05:44
I can't unfortunately, to my knowledge the hard drive onboard the AC only cache's so many logs, since the time of the incident, I did turn on the device a few times trying to investigate some things, so got rid of the older records, unless I'm wrong?

I'm not sure what the Spark's storage capacity is. But, the P3 and Mavic are capable of storing several hundred minutes. I'd be surprised if the Spark isn't the same. Please take another look to see if those requested .DATs are there. BTW, you can select more than one flight in the DJI Assistant to be included in the DJI_ASSISTANT_EXPORT_FILE_XXXXXX.DAT. You can then extract them by using ExtractDJI.
2017-10-6
Use props
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

BudWalker Posted at 2017-10-6 05:59
I'm not sure what the Spark's storage capacity is. But, the P3 and Mavic are capable of storing several hundred minutes. I'd be surprised if the Spark isn't the same. Please take another look to see if those requested .DATs are there. BTW, you can select more than one flight in the DJI Assistant to be included in the DJI_ASSISTANT_EXPORT_FILE_XXXXXX.DAT. You can then extract them by using ExtractDJI.

Unfortunately, it seems it really does store only a limited amount of data and flight logs. I have nothing before the 1st of October. The only thing I may be able to do recover is pulling the flight log from the phone itself, but this is not as granular so not sure how much insight it will provide?

DJI

DJI
2017-10-6
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline

bfreedman Posted at 2017-10-6 06:16
Unfortunately, it seems it really does store only a limited amount of data and flight logs. I have nothing before the 1st of October. The only thing I may be able to do recover is pulling the flight log from the phone itself, but this is not as granular so not sure how much insight it will provide?

Bummer.

While you're at it could you select all the flights and provide a Dropbox link to the generated DJI_EXPORT_XXXX.DAT. There could be something interesting in one of the included .DATs

I looked at the two .DATs you provided today, FLY060 and FLY069. I noticed that there was a compass calibration after the incident flight and before FLY060. Had you done a calibration prior to the incident flight?
2017-10-6
Use props
bfreedman
lvl.1
Flight distance : 64849 ft
Australia
Offline

Unfortunately the flights between FLY060 and FLY069 are meaningless, in the sense I just turned on the AC a few times to check a few things, but never actually flew. That’s the reason why I included these files because it’s the best data points I have! Bummer - yes! I however can confirm that yes, I DID do a compass calibration, just after the incident (therefore before FLY060). How did you determine this, or just assumed because Yaw and Mag Yaw seem a little closer now?
2017-10-6
Use props
djiuser_Kf4iPA3
First Officer
Flight distance : 510968 ft
Israel
Offline

SparkeD Posted at 2017-10-5 06:10
I have experienced this ATTI mode, no GPS signal and erratic flight response twice now. Once in August when flying above a lake and just yesterday when flying neat a bike trail.  The drone seems to get internally switched to sport mode and does not respond to remote control commands.  Luckily the Spark only clipped a tree and I was able to regain control after a short period of uncontrolled flight. This is very nerve racking.

Am I the only one seeing a contradiction here?
in the second message from the bottom, the app advises the user to "please switch to atti mode if craft behave abnormally"...
but... the user can't do that, right? you can't switch to ATTI mode manually... so what is this advice doing here?
2017-10-6
Use props
fansbdb5997e
New

Malaysia
Offline

My DJI GPS module had cracked.. I also not sure why that will happen.. so can anyone tell me where can I buy the DJI spark GPS module...
2017-10-6
Use props
hallmark007
Captain
Flight distance : 9812789 ft
  • >>>
Ireland
Offline

djiuser_Kf4iPA3 Posted at 2017-10-6 08:36
Am I the only one seeing a contradiction here?
in the second message from the bottom, the app advises the user to "please switch to atti mode if craft behave abnormally"...
but... the user can't do that, right? you can't switch to ATTI mode manually... so what is this advice doing here?

Ok fair point it should be updated, it is the same message you receive on Mavic and phantom, although on phantom you have a switch, but all new models from dji are now auto switch to Atti , Spark Mavic phantom 3,4,4advanced and Pro inspire .
It doesn’t make any difference as aircraft will auto switch to Atti , but yes maybe it should say aircraft switching to Atti.
2017-10-6
Use props
djiuser_Kf4iPA3
First Officer
Flight distance : 510968 ft
Israel
Offline

hallmark007 Posted at 2017-10-6 11:29
Ok fair point it should be updated, it is the same message you receive on Mavic and phantom, although on phantom you have a switch, but all new models from dji are now auto switch to Atti , Spark Mavic phantom 3,4,4advanced and Pro inspire .
It doesn’t make any difference as aircraft will auto switch to Atti , but yes maybe it should say aircraft switching to Atti.

Fair enough, I wasn't aware of this capability in other drones. Since GO4 applies to an entire family of drones and not just Spark, I guess it makes sense.
2017-10-6
Use props
BudWalker
Captain
Flight distance : 5426050 ft
  • >>>
United States
Offline

bfreedman Posted at 2017-10-6 07:08
Unfortunately the flights between FLY060 and FLY069 are meaningless, in the sense I just turned on the AC a few times to check a few things, but never actually flew. That’s the reason why I included these files because it’s the best data points I have! Bummer - yes! I however can confirm that yes, I DID do a compass calibration, just after the incident (therefore before FLY060). How did you determine this, or just assumed because Yaw and Mag Yaw seem a little closer now?

The eventLog stream has the bias and scale values for the magnetometers. These will change when a compass calibration is done. From FLY058

-66.452 :[L-COMPASS][read app cali data]fmu:0,user:0,scale:[ 2.613  2.525  2.678]
-66.452 :[L-COMPASS][read app cali data]fmu:0,user:0,bias:[557.127 -1341.214 -1033.606]


And from FLY060

-87.101 :[L-COMPASS][read app cali data]fmu:0,user:0,scale:[ 2.432  2.832  2.402]
-87.101 :[L-COMPASS][read app cali data]fmu:0,user:0,bias:[921.769 -1396.186 -1054.675]


I suppose you're thinking that the compass calibration fixed the problem. It's hard to say. That's why I was looking for the flights previous to the incident flight. There have been a few incidents with the P4 similar to yours. In each case the incident flight did not follow a compass calibration, in fact it had been many flights since a calibration. The previous flights didn't have problems suggesting there was an abrupt change in the compass hardware. With one of those incidents there was subsequent testing before a calibration was done. In that case it was clear the calibration fixed the problem. The others, not so clear.

You could take a look at those incidents.
https://phantompilots.com/threads/fly-away.117049/page-2#post-1207461

https://phantompilots.com/thread ... 23179/#post-1257913

2017-10-6
Use props
DJI Susan
Administrator
Offline

djiuser_xLE6O03 Posted at 2017-9-30 02:55
2. no error message before take off .
here is xls file from csv , i colored the part wit the problem https://www.dropbox.com/s/oyye9wuxdulba8n/FLY056_color.xlsx?dl=0
here is scren record of playback log in app

Thanks for getting back to us. I'm sorry to tell you that we do not support the third party software. If you have any doubts of the drone, you are welcome to send it back for fully check.
2017-10-9
Use props
Mishanki
lvl.4
Flight distance : 1240358 ft
Russia
Offline

I still have problem with compass: airdate, black box. Almost every day. Sometimes mistakes happen far in the field. Sometimes not far from me. Spark rotates a little counter-clockwise or GPS Weak signal. Turns if it's a long time to fly straight.
I calibrated in the field repeatedly.
p.s. 2. no error message before take off . - same.


2018-1-4
Use props
Wachtberger
Captain
Flight distance : 261509 ft
Germany
Offline

Mishanki Posted at 2018-1-4 07:14
I still have problem with compass: airdate, black box. Almost every day. Sometimes mistakes happen far in the field. Sometimes not far from me. Spark rotates a little counter-clockwise or GPS Weak signal. Turns if it's a long time to fly straight.
I calibrated in the field repeatedly.
p.s. 2. no error message before take off . - same.

Would you mind uploading the flight record here http://www.phantomhelp.com/LogViewer/Upload/ and share the link with us? I have an initial idea but need to verify something first.
2018-1-4
Use props
Mishanki
lvl.4
Flight distance : 1240358 ft
Russia
Offline

Wachtberger Posted at 2018-1-4 07:43
Would you mind uploading the flight record here http://www.phantomhelp.com/LogViewer/Upload/ and share the link with us? I have an initial idea but need to verify something first.

http://www.phantomhelp.com/LogViewer/FLNBCEVWAWH0B3GNWT4F/ - today
http://www.phantomhelp.com/LogViewer/9L2P6EAB4XWF54U1QTJT/ - first flight
http://www.phantomhelp.com/LogViewer/15V0YY3MWGPQWNNDIDC5/ - 4 flight
http://www.phantomhelp.com/LogViewer/BGHG89P26SA66Z9TSPYK/ - again
http://www.phantomhelp.com/LogViewer/M160JVEMHDZQHKYD3AN4/ - some one

... and many problems with stable horizont and straight direction. Example with bad direction:

DJI_0084.jpg
2018-1-4
Use props
Wachtberger
Captain
Flight distance : 261509 ft
Germany
Offline


Ok, I can see that all flights are in the same area. Have you ever tried flying somewhere else and have had the same problems? There could be a source of strong interferences in that area. Or do you carry any bluetooth device with you?
To be honest I cannot see more but will ask a much more experienced user to have a look at your flightlogs.
In any case I am happy that each time you have managed to regain control of your Spark and land it safely.
P.S. The video doesn't work.

P.P.S. Now the video works.
2018-1-4
Use props
123Next >
Advanced
You need to log in before you can reply Login | Register now

Credit Rules