Got this GPS message today

[sar104]

I sometimes wonder how my pockets arent stripped of their metal contents when I walk on concrete or next to metal things. These things dont usually affect my old school pocket compass, or the fancy digital one in the mirror of my car. I cant help but think there isnt a little crying wolf going on with this mavic compass stuff.

You might have to elaborate on where you were going with that post, but I'll take a stab at addressing what you might have been trying to say. The earth's magnetic field is quite weak. You don't notice it grabbing ferromagnetic materials that you might be carrying. As a result, it does not require a particularly strong magnetic field to disrupt a measurement of the earth's magnetic field. As for your pocket compass, I can only assume that you have never tested it within a few inches of the surface of rebar-loaded concrete, or the steel panels of a vehicle. If you did you would find that it was significantly affected.

More specifically, if you were to examine the raw magnetometer data contained in the DAT files of UAVs that have experience magnetic interference, then you would not even need to speculate on whether these effects are real - you would be able to observe direct measurements of the effects. I'm guessing that you haven't done that either.

 

[ac0j]

You would be correct on all parts. BUT, I might take my pocket compass out fro some experiments when its not so cold outside.

So you are saying that the dat files show tons of compass anomalies, or just when the compass errors are observed? Not all metal is magnetized.

 

[sar104]

You would be correct on all parts. BUT, I might take my pocket compass out fro some experiments when its not so cold outside.

So you are saying that the dat files show tons of compass anomalies, or just when the compass errors are observed? Not all metal is magnetized.

Do some tests. You will be surprised, apparently. The DAT files for flights with compass and yaw errors show the problem very clearly, because they contain the raw magnetometer readings. Flights with no errors show few anomalies - not surprising since it is the anomalies that trigger the errors - specifically discrepancies between the IMU heading and the magnetic heading.

 

[ac0j]

I find it amazing. I cant believe that 4 high current motors running at constantly varying frequencies, the wires connecting those to the battery, and all the currents flowing INSIDE of the the mavic (along with the compass) dont disrupt the compass, but the static rebar in my driveway could?
I am just saying I believe there are compass errors. No doubt there is from time to time. But, I think the take off point, and where and when you calibrate the compass has less to do with it than other factors like Radio towers near the flight, high tension power lines, and things like that. Hence my crying wolf statement.

 

[sar104]

I find it amazing. I cant believe that 4 high current motors running at constantly varying frequencies, the wires connecting those to the battery, and all the currents flowing INSIDE of the the mavic (along with the compass) dont disrupt the compass, but the static rebar in my driveway could?
I am just saying I believe there are compass errors. No doubt there is from time to time. But, I think the take off point, and where and when you calibrate the compass has less to do with it than other factors like Radio towers near the flight, high tension power lines, and things like that. Hence my crying wolf statement.

Fair enough - but I would have to respond that, based on that statement, your understanding of the underlying physics is negligible. Radio frequency emissions have no effect on these kinds of magnetometers - they do not produce any measurable magnetic field. AC transmission lines are high-voltage, low-current systems and, since the magnetic field around a conductor is proportional only to current, also have little effect. Plus you don't take off within inches of such sources. Ferromagnetic objects, on the other hand, such as rebar and vehicles, produce a significant local field that is trivial to measure. And yet you are touting your "belief" that this is not important? What you might "believe" is entirely irrelevant to anyone but yourself.

 

[ac0j]

Fair enough - but I would have to respond that, based on that statement, your understanding of the underlying physics is negligible. Radio frequency emissions have no effect on these kinds of magnetometers - they do not produce any measurable magnetic field. AC transmission lines are high-voltage, low-current systems and, since the magnetic field around a conductor is proportional only to current, also have little effect. Plus you don't take off within inches of such sources. Ferromagnetic objects, on the other hand, such as rebar and vehicles, produce a significant local field that is trivial to measure. And yet you are touting your "belief" that this is not important? What you might "believe" is entirely irrelevant to anyone but yourself.

You are obviously a very smart person. I agree you are way smarter than me. But you are side stepping part of my point.
The MAVIC is DC. The motor wires would absolutely have an affect on my mechanical compass. The pionter would align directly along the length of the wire. No matter where north is. THAT force is MUCH stronger than the rebar, or metal structures, and you cant get them much closer to the compass assembly in the mavic. If the internal compass can shield the effect of what is going on inside the craft, I would expect it to easily overcome the near immeasurable magnetic influence of rebar, or metal objects.
My comments are in reaction to all the compass calibration discussion that are implying that calibration is causing inflight compass errors.
But then again, I am a dumb guy, and must yield to input from the lofty crows nests of the educated.

 

[sar104]

You are obviously a very smart person. I agree you are way smarter than me. But you are side stepping part of my point.
The MAVIC is DC. The motor wires would absolutely have an affect on my mechanical compass. The pionter would align directly along the length of the wire. No matter where north is. THAT force is MUCH stronger than the rebar, or metal structures, and you cant get them much closer to the compass assembly in the mavic. If the internal compass can shield the effect of what is going on inside the craft, I would expect it to easily overcome the near immeasurable magnetic influence of rebar, or metal objects.
My comments are in reaction to all the compass calibration discussion that are implying that calibration is causing inflight compass errors.
But then again, I am a dumb guy, and must yield to input from the lofty crows nests of the educated.

What is the quantitative basis for asserting that the current in the motor supply is stronger than external sources? And I pose that question in isolation from the obvious observation that it clearly isn't, otherwise these compasses would never work properly. And the question is also rhetorical, because there is no such basis. The motor currents are not only routed to avoid the vicinity of the compasses but the return current is virtually coaxial with the supply current, and so the net magnetic field from those conductors is approximately zero.

So I apologize for being so blunt but I find it frustrating that so many people feel justified in making completely unsupported assertions on technical subjects from positions of almost total ignorance, and then attempt to spin it as some kind of unreasonable elitism when they are called on it.

 

[Spy808]

No worries... I'm in the testing phase for the fix on the GPS\Çompäss module & front compass board... There should be no hassles calibrating module if it was working correctly to begin with... No Cfixer bs & do this do that kind of thing... It should work on the phanthom drones too... Just trÿïn to figure out how to make a stööpïd YouTube video on how n whÿ to fix the GPS\Compass...

 

[ac0j]

What is the quantitative basis for asserting that the current in the motor supply is stronger than external sources? And I pose that question in isolation from the obvious observation that it clearly isn't, otherwise these compasses would never work properly. And the question is also rhetorical, because there is no such basis. The motor currents are not only routed to avoid the vicinity of the compasses but the return current is virtually coaxial with the supply current, and so the net magnetic field from those conductors is approximately zero.

So I apologize for being so blunt but I find it frustrating that so many people feel justified in making completely unsupported assertions on technical subjects from positions of almost total ignorance, and then attempt to spin it as some kind of unreasonable elitism when they are called on it.

First off. I have said multiple times you are smarter than me. So, let’s get this part out of the way.
I respect what you post here. I believe every technical thing you say. Furthermore, as you have noticed and affirmed, I may be one of the top ten dumbest people that you will ever come across.
There, now that I have submitted and your foot is firmly planted on my throat, use your power to teach instead of scorn and ridicule.

I had typed the reworded question here but I deleted it.

You know what? Never mind there isn’t but 1 or 2 people here that would understand the answer anyway.

 

[sar104]

First off. I have said multiple times you are smarter than me. So, let’s get this part out of the way.
I respect what you post here. I believe every technical thing you say. Furthermore, as you have noticed and affirmed, I may be one of the top ten dumbest people that you will ever come across.
There, now that I have submitted and your foot is firmly planted on my throat, use your power to teach instead of scorn and ridicule.

please enlighten us how when calibrating the compass the minuscule amounts of magnetic energy effects the process without generating a calibration error. And how the shielding is only good for the energy contained within the Mavic, and not it’s surroundings and how compass errors are not a defective Mavic.
When I calibrate my Mavic ina bad spot it says unsuccessful and I have to move and do it again.

You know what? Never mind there isn’t but 1 or 2 people here that would understand the answer anyway.

There you go again. Post #87 answers the question about motor currents in the simplest possible terms and yet you ignore that answer completely and at the same time both demand to be educated and object to any attempt to educate. I cannot explain this any more clearly.

The rebar in concrete, by contrast, can locally significantly distort the earth's magnetic field. You can measure it. It's that simple. It can result in spurious initial magnetic heading readings which, in turn, leads to subsequent disagreement between the IMU heading and the magnetic heading and GPS track. There is nothing mysterious about this at all.

The rest of your question doesn't even make sense. What energy? What shielding? What does the fact that the FC can often detect a bad calibration attempt have to do with the effects of magnetic interference at take off on the FCs sensor fusion scheme? There is no point asking technical questions if you have no interest in reading the answers or even considering modifying your understanding based on those answers. Either make some kind of attempt to understand or, if you can't be bothered, just write it off as beyond your interests. But posting gut feelings or beliefs about the interactions of various magnetic fields is completely pointless - that's not how physics works.

 

[BudWalker]

@ac0j I responded to your related question in post #57. But, I didn't see any responses. Does that answer work for you? If not, I can clarify further.

Responding to your post #66. The motor wires inside the Mavic are twisted which has the effect of cancelling the magnetic fields due to the DC current in each wire. Just like Ethernet cable has twisted pairs for much the same reason. In fact, at one time the Mavic had a problem with this. Check out

DIY rear compass cabling fix

For sure, the engineering that goes into attenuating extraneous magnetic fields is a complicated, detailed business. But, DJI seems to have figured it out, except for the manufacturing problem referenced above.

 

[ac0j]

So it is the experts opinion that calibrating the compass ona driveway causes the in flight compass errors? That was the point/question I had with my post “crying wolf” that started all the chest pounding. What about the actual magnets I n the motors? I was just surprised that the rebar theory is more of an issue than the Mavic internals. That is all. Really didn’t ask for an explanation, just posted a thought at first. I hope other compass dependent systems are better at ignoring common magnetic behavior.

Hey, this is mainly just giving another chance to be called an idiot! Seems to enjoy that.

 

[sar104]

So it is the experts opinion that calibrating the compass ona driveway causes the in flight compass errors? That was the point/question I had with my post “crying wolf” that started all the chest pounding. What about the actual magnets I n the motors? I was just surprised that the rebar theory is more of an issue than the Mavic internals. That is all. Really didn’t ask for an explanation, just posted a thought at first. I hope other compass dependent systems are better at ignoring common magnetic behavior.

Hey, this is mainly just giving another chance to be called an idiot! Seems to enjoy that.

Actually I think the consensus is that taking off from a magnetically distorted location is much more of a problem than calibrating in such a location because, generally, the calibration will fail. Bad calibrations were common back in the P2 days but, as @BudWalker has argued, the newer FCs appear to be much better at detecting calibration problems.

And it was your post #64, that was not just about calibration, that specifically started the discussion:

"I find it amazing. I cant believe that 4 high current motors running at constantly varying frequencies, the wires connecting those to the battery, and all the currents flowing INSIDE of the the mavic (along with the compass) dont disrupt the compass, but the static rebar in my driveway could?
I am just saying I believe there are compass errors. No doubt there is from time to time. But, I think the take off point, and where and when you calibrate the compass has less to do with it than other factors like Radio towers near the flight, high tension power lines, and things like that. Hence my crying wolf statement."​

 

[BudWalker]

So it is the experts opinion that calibrating the compass ona driveway causes the in flight compass errors? That was the point/question I had with my post “crying wolf” that started all the chest pounding. What about the actual magnets I n the motors? I was just surprised that the rebar theory is more of an issue than the Mavic internals. That is all. Really didn’t ask for an explanation, just posted a thought at first. I hope other compass dependent systems are better at ignoring common magnetic behavior.

Hey, this is mainly just giving another chance to be called an idiot! Seems to enjoy that.

While it's possible to obtain a flawed calibration that will cause in flight problems it almost never happens. You may have read the other posts here where this was discussed. If you attempt to calibrate at a bad location the "interference detetced, move to a better location" message will be issued and the calibration rejected. I suspect that part of the reason that some believe in the flawed-calibration-causes-erratic-flight scenario is that because it's similar to the launch-from-geomagnetically-distorted-site scenario.

Another thing is that the Mavic internals rotate with the Mavic but the rebar doesn't. This is a crucial difference because magnetic effects that rotate with the AC can be detected during the calibration process and then compensated for. Magnetic effects that don't rotate with the AC can be detected but the data is ambiguous and can't be compensated for.

The motor magnets cause what are called "hard iron" distortions. These are detected during the calibration process. The hard iron effects are then compensated for as part of the calibration process.

 

[ac0j]

While it's possible to obtain a flawed calibration that will cause in flight problems it almost never happens. You may have read the other posts here where this was discussed. If you attempt to calibrate at a bad location the "interference detetced, move to a better location" message will be issued and the calibration rejected. I suspect that part of the reason that some believe in the flawed-calibration-causes-erratic-flight scenario is that because it's similar to the launch-from-geomagnetically-distorted-site scenario.

Another thing is that the Mavic internals rotate with the Mavic but the rebar doesn't. This is a crucial difference because magnetic effects that rotate with the AC can be detected during the calibration process and then compensated for. Magnetic effects that don't rotate with the AC can be detected but the data is ambiguous and can't be compensated for.

The motor magnets cause what are called "hard iron" distortions. These are detected during the calibration process. The hard iron effects are then compensated for as part of the calibration process.

Good post! Very helpful and now I understand.
It’s nice when smart people take the time to enlighten others instead of constantly reminding us that we aren’t as smart as them. I didn’t get to 55 years old without encountering many people like that. It’s jyst the way things are, and I have learned to expect it.

I entered the conversation after reading many times here that you should never calibrate the compass because if you do it will cause problems. I don’t believe that to be true in most cases. I feel like I am not alone now.

 

[Jenningsfarm]

View attachment 28295 So today I was coming in when I got the low battery warning. And I was not far away. The drone was right in front of me and there were obviously plenty of satellites in range. And I had been flying back-and-forth in this area with no issues before. And all of a sudden I got this message saying the GPS signal was gone. Luckily the drone was right next to me otherwise I would have been terrified. Any ideas why this would happen when it clearly shows I have lots of satellites. Some reason it went in atti mode. But why when there were 18 satellites? Also I should mention that the signal came back a minute or so later and went back to GPS mode.

Hi i notice the focus square is on mine allso did this 20 mtrs above no interference heaps of satellites i think the focus may have interfered.

 

[Andyc]

I used to panic a little over compass calibrations due to reading this forum, but having purchased a Mavic, calibrated the compass before the first flight and flown it many times now without even a hint of an error?....... I do take a look every now and again to check (the two) compass indicators and they are always in the green!........ I have taken the Mavic through airport scanners and travelled 3000mls and when I get it out of the bag it seems perfectly happy to fly just the same as it does in my home county?........ I'm still not sure what's right and what's wrong but I do know what works for me right now?.......

 

[ac0j]

When my mavic asks for a calibration, I do it. It doesnt happen very often. If my magnetic driveway or car is affecting the calibration, it tells me to move and do it again. compass errors happen, dont get caught up in the where and when to calibrate your compass debate. That mess is more about who is smarter than who.

 

[Dabrain]

I wouldn't worry about it unless it happens again. Take it out to somewhere open and safe to fly, like a big field or parking lot. Do some test flying, keeping it close and relatively low, progressively go further and higher and come back, do some close flying in Sport mode too.

If the error doesn't reoccur and everything works well, you can chalk it up to some anomalous magnetic field of big hunk of metal nearby, and carry on a happy pilot.

Don't calibrate the compass unless GO4 tells you to.

Naw actually DJI now suggests:
the compass be calibrated when:

• Flying at a location further than 10 km away from the last flight location
• The Mavic has not been flown for 1+ months
• A compass error is displayed in DJI GO and the error remains after changing locations

 

[Dabrain]

He is exactly right. DO NOT calibrate your compass before every flight. Do it once and leave it alone.

DJI is now recommending the compass be calibrated when:

• Flying at a location further than 10 km away from the last flight location
• The Mavic has not been flown for 1+ months
• A compass error is displayed in DJI GO and the error remains after changing locations