Mavic 2 Pro Flyaway

[sar104]

Just returned from a small trip in a city 500Km away my home town. The very first moment I tried to fly my MP2 in that city it requested a compass calibration. I did several flights in the area and there was no other request for compass calibration.

When I returned back, I tried to fly in my neighborhood. In that moment the MP2 asked for a calibration again. All the next flights did not request a calibration.

I am at the latest version in both ios app and firmware, therefore I believe the 50Km distance for triggering the calibration is still present.

I wish DJI could explain why they decided to put in the new Mavic 2 such a procedure when in the old MP1 there wasn't.

DJI typically doesn't explain these things at all. It would be interesting to know exactly what their reasoning was though - they must have felt that there was some advantage to making that change.

 

[peros550]

DJI typically doesn't explain these things at all. It would be interesting to know exactly what their reasoning was though - they must have felt that there was some advantage to making that change.

Yes I agree with you on that. What is really strange is that there are reports of people who were never requested to re-calibrate compass and they have definitely flown to places farther than 50km

 

[sar104]

Yes I agree with you on that. What is really strange is that there are reports of people who were never requested to re-calibrate compass and they have definitely flown to places farther than 50km

Yes - I haven't seen any DAT files from those aircraft, but I have seen them from aircraft that were not requesting 30 day calibrations. In those cases the "last calibration" date was years in the future. I've no idea how or why that was the case.

There have been a few reports that these were removed in the last firmware update, but I'm still getting 30 day requests, so it may be that those reports are still outliers.

 

[kilomikebravo]

Thanks, SAR. Obviously, DJI thinks (or thought) that forcing a cal after the aircraft is moved to a new location some distance away was a good idea although, it doesn't make a lot of sense. Mine did right after the first 30 days but since the later firmware was installed, no requests have been triggered due to any time limit but, I haven't moved it more than 50 miles either.

 

[sar104]

Thanks, SAR. Obviously, DJI thinks (or thought) that forcing a cal after the aircraft is moved to a new location some distance away was a good idea although, it doesn't make a lot of sense. Mine did right after the first 30 days but since the later firmware was installed, no requests have been triggered due to any time limit but, I haven't moved it more than 50 miles either.

Interesting. There are various possibilities - perhaps it's just a difference of opinion between different DJI engineers, perhaps they have seen a lot of Mavics with compass issues caused during travel/transport, perhaps it's related to more people adding accessories that affect the aircraft's magnetic field.

 

[jcampen]

jcampen: We all understand your frustration but in this case, WTB is correct, you do not need to calibrate the compass simply because the aircraft is moved from its previous location, regardless of distance. Ever tried using the compass on your phone? You can fly across the globe and it will still give you accurate information as long as you're not in an area with magnetic interference. The drone works the same way. AFAIK, every instance reported was caused by local magnetic interference. If the drone has a good heading and good GPS, with no other issues, it will fly normally.

I dont disager K , its my own view as well. Then perhaps its a DJI software requirement?. I have been to China 7 times in 6 months, Australia once, Thailand once and Korea a few times as well and back to Malaysia in between most of those trips. There is nearly never one time I dont need to do a compass calibration before I can fly those distances. I even try moving it around to different spots, but until I do the calibration, nothing works. Its does seem however this deosnt affect the P4P as much.

 

[paulatkin73]

Just returned from a small trip in a city 500Km away my home town. The very first moment I tried to fly my MP2 in that city it requested a compass calibration. I did several flights in the area and there was no other request for compass calibration.

When I returned back, I tried to fly in my neighborhood. In that moment the MP2 asked for a calibration again. All the next flights did not request a calibration.

I am at the latest version in both ios app and firmware, therefore I believe the 50Km distance for triggering the calibration is still present.

I wish DJI could explain why they decided to put in the new Mavic 2 such a procedure when in the old MP1 there wasn't.

So, there is a possible explanation here. There is such a thing as magnetic declination. It differs quite significantly if you travel far enough, and, say, arducopter code retrieves it from a dictionary table based on gps coordinates. But it can be that dji logic demands recalibration if new declination amount differs enough. Or, if drone during transportation was near to something magnetic, like, woofer speakers in the trunk. It is also possible.

 

[sar104]

I dont disager K , its my own view as well. Then perhaps its a DJI software requirement?. I have been to China 7 times in 6 months, Australia once, Thailand once and Korea a few times as well and back to Malaysia in between most of those trips. There is nearly never one time I dont need to do a compass calibration before I can fly those distances. I even try moving it around to different spots, but until I do the calibration, nothing works. Its does seem however this deosnt affect the P4P as much.

It's a requirement implemented in the M2 firmware. As soon as the aircraft has acquires its location it checks to see how far that is from the location of the last compass calibration. If it is more than 50 km then it requires a recalibration. It's simply an arbitrary request - it is not based on any direct compass measurements.

 

[sar104]

So, there is a possible explanation here. There is such a thing as magnetic declination. It differs quite significantly if you travel far enough, and, say, arducopter code retrieves it from a dictionary table based on gps coordinates. But it can be that dji logic demands recalibration if new declination amount differs enough. Or, if drone during transportation was near to something magnetic, like, woofer speakers in the trunk. It is also possible.

Compass calibration cannot measure and doesn't compensate for changes in declination and, as I mentioned above, the requirement is simply based on the distance from the previous calibration. If the magnetic characteristics of the aircraft had changed during transport then that would be a good reason to recalibrate - so perhaps that's what DJI is trying to deal with proactively.

 

[kilomikebravo]

SAR: I'd bet a fair amount of money that you are right on the money with that guess. Avoid troubleshooting calls if possible, makes perfect sense.

 

[Anterialis]

I've read through most of the posts in this thread with great interest. In my opinion, one thing is to help the OP, another great reason for posting here is for educating other pilots (like myself). I thank you all greatly for putting in the efforts and thorough explanations.

I don't mean to restart this thread, simply I would like to ask a question about the terminology used and the science behind it. I believe it could help others and definitely myself in undestanding this complex albeit "easy" subject.

Have I understood things correctly if I say that:

1. magnetic declination is the difference between true north and magnetic north?
2. magnetic deviation is the difference between actual magnetic north and the "north" you get if the compass is affected by nearby metallic objects? IOW the correction for "false magnetic inputs" created by the (in this example) mavic pro itself?
3. a good "compass calibration" is actually just something you have done if you have corrected for the magnetic deviation?
4. The compass (obviously) gives data about the magnetic north (not true north), and the Mavic Pro has a "look up table" to find magnetic declinations for different (probably thousands or billions??) places on earth. Thus you don't need to "recalibrate" the compass ever if once done poperly and the magnetic field created by the drone itself has not changed?
5. and this all leads me to my claim: IF you have done a proper "compass calibration" and thus corrected for the magnetic deviation (induced by the drone itself) AND the magnetic field in the drone has not changed, in theory, you should never ever need to "re-calibrate" the compass?
6. I also claim that one should actually NOT recalibrate the compass often as it will probably one time not only correct for the drone's magnetic field but also e.g. metall in the concrete underneath the drone, and thus the compass will give erroneous data after takeoff (because you told the drone by doing a compass calibration that magnetic north must correct for the magnetic field in the drone AND from the ground, and you left the ground after takeoff...)?

7) Also, can the magnetic field in the drone change if you do not add items to it? E.g. if I were to put the mavic next to a strong magnet for hours and then remove the magnet, would the magnetic field in the mavic have changed? Will it return to "normal" after a while?

Sorry for my difficult english and many questions, but hopefully - as I mentioned - this can add to the understanding of this subject for me and others. Please correct me anywhere I am wrong! And thanks again for the tremendous efforts put in by all of you, especially thanks to sar104 and Meta4.

 

[sar104]

Have I understood things correctly if I say that:
magnetic declination is the difference between true north and magnetic north?

Yes.

magnetic deviation is the difference between actual magnetic north and the "north" you get if the compass is affected by nearby metallic objects? IOW the correction for "false magnetic inputs" created by the (in this example) mavic pro itself?

No - magnetic deviation is a synonym for magnetic declination.

a good "compass calibration" is actually just something you have done if you have corrected for the magnetic deviation?

Compass calibration has nothing to do with declination/deviation - it is the measurement of the magnetic field of the aircraft itself so that it can be subtracted from the total measured field to leave just the earth's magnetic field.

The compass (obviously) gives data about the magnetic north (not true north), and the Mavic Pro has a "look up table" to find magnetic declinations for different (probably thousands or billions??) places on earth. Thus you don't need to "recalibrate" the compass ever if once done poperly and the magnetic field created by the drone itself has not changed?

Yes. But it's not a lookup table - it's a computational model.

and this all leads me to my claim: IF you have done a proper "compass calibration" and thus corrected for the magnetic deviation (induced by the drone itself) AND the magnetic field in the drone has not changed, in theory, you should never ever need to "re-calibrate" the compass?

That is the logical conclusion.

I also claim that one should actually NOT recalibrate the compass often as it will probably one time not only correct for the drone's magnetic field but also e.g. metall in the concrete underneath the drone, and thus the compass will give erroneous data after takeoff (because you told the drone by doing a compass calibration that magnetic north must correct for the magnetic field in the drone AND from the ground, and you left the ground after takeoff...)?

That shouldn't happen because, in the calibration process, in the frame of reference of the aircraft, there is a constant magnetic field (the aircraft) and a varying magnetic field (the earth's plus any local distortion). The process only subtracts the constant field, and so should not be affected by any external field. That's clearly not true though for strong enough external fields. So, on balance, I agree that it is undesirable to do frequent, unnecessary calibrations.

Also, can the magnetic field in the drone change if you do not add items to it? E.g. if I were to put the mavic next to a strong magnet for hours and then remove the magnet, would the magnetic field in the mavic have changed? Will it return to "normal" after a while?

The aircraft has ferromagnetic materials, and so they are capable of acquiring permanent magnetism if exposed to strong enough magnetic fields.

 

[Anterialis]

Thank you so much, sar104. I cant begin to tell you how greatful I am for your contributions on this forum =)

About deviation. I believe I learned (in Norway during my sailing course) that declination is the difference between T and M north, and that deviation is the error caused by local magnetic fields. I do not mean to question your knowledge (which is far superior to mine...), but are you sure that these two terms are not different? I know wikipedia can't be trusted 100%, but in the norwegian and english version of the article "magnetic deviation" it states that "Magnetic deviation is the error induced in a compass by local magnetic fields, which must be allowed for, along with magnetic declination, if accurate bearings are to be calculated. (More loosely, "magnetic deviation" is used by some to mean the same as "magnetic declination".".

Can you say a few words on the latter?

 

[sar104]

Thank you so much, sar104. I cant begin to tell you how greatful I am for your contributions on this forum =)

About deviation. I believe I learned (in Norway during my sailing course) that declination is the difference between T and M north, and that deviation is the error caused by local magnetic fields. I do not mean to question your knowledge (which is far superior to mine...), but are you sure that these two terms are not different? I know wikipedia can't be trusted 100%, but in the norwegian and english version of the article "magnetic deviation" it states that "Magnetic deviation is the error induced in a compass by local magnetic fields, which must be allowed for, along with magnetic declination, if accurate bearings are to be calculated. (More loosely, "magnetic deviation" is used by some to mean the same as "magnetic declination".".

Can you say a few words on the latter?

You are completely correct - I was thinking of magnetic variation, which is another term for magnetic declination. Magnetic deviation is the term for local magnetic distortions.

 

[Meta4]

About deviation. I believe I learned (in Norway during my sailing course) ...

If you owned a sailboat, you may have used a compass adjuster to correct the magnetic deviation affecting your boat's compass.
The compass adjuster would place small magnets fore and aft and left and right of the compass to neutralise the boat's own magnetic fields so that the compass reading was accurate.

Think of calibrating the drone's compass as the digital equivalent of calling in the compass adjuster.
It measures the drone's own field so that it can be subtracted from the total magnetic field the compass detects.
That gives an accurate compass reading, not distorted by any magnetism that's part of the Mavic.

 

[Anterialis]

Exactly. I also learned what was said about static and dynamic magnetic fields and that the drone subtracts the dynamic field from its calculations when “calibrating” the compass. That was new and a useful piece of information to me to better understand things. Thank you both for what you bring to the table

 

[JakeBlues]

Hi, let me just drop in a small question to you here .... and it might turn out just nothing BUT, are you using PolarPro ND filters with their “magnetic” casing? There is ample proof that, if you transport that casing anywhere near the drone, it upsets the magnetic field “enough” to trigger a Re-calibration request anytime you change filters or just have the box nearby. This was also reported by several reputable YouTubers ! Check it out!

 

[kilomikebravo]

There is ample proof that, if you transport that casing anywhere near the drone, it upsets the magnetic field “enough” to trigger a Re-calibration request

Jake: FWIW, I keep four PP filter cases in my backpack along with the M2 and I've never had a compass cal request triggered.

 

[Anterialis]

Hi, let me just drop in a small question to you here .... and it might turn out just nothing BUT, are you using PolarPro ND filters with their “magnetic” casing? There is ample proof that, if you transport that casing anywhere near the drone, it upsets the magnetic field “enough” to trigger a Re-calibration request anytime you change filters or just have the box nearby. This was also reported by several reputable YouTubers ! Check it out!

Now this would probably be a good reason to actually do a compass calibration even though the drone does not request it as you might be changing (albeit only a little bit) the static magnetic field on the drone. If the drone does not request it though, it might be too small to make a difference...

 

[JakeBlues]

Now this would probably be a good reason to actually do a compass calibration even though the drone does not request it as you might be changing (albeit only a little bit) the static magnetic field on the drone. If the drone does not request it though, it might be too small to make a difference...

Hi again,

Just because “others” didn’t “yet” experience interference doesn’t mean it’s not there, so just err on the side of caution with PP filter boxes that are the latest with magnetic closing mechanism.

Please note, I am using only PP products and became aware by researching the new 6-pack for M2P and Zoom.

Here the link:

If you like to see the imperative part forward the clip to about 6:15

Best regards
Jake