Mavic Pro behaving strangely
- Thread starter Jackcutrone
- Start date
Got that.....but wouldn't the AC and the RC display not be in agreement as to direction? or are we both saying the same thing?
Don't think you are missing anything. It's only when the aircraft and the display are different that you have a problem, at least as i understand it.
Yes, the AC if physically facing north wouldn't be showing to face north in the app. But since the interference is attached to the AC, moving the AC won't help. Instead you'd need to do a calibration.
Now if you happen to have interference from the Trackmo and with whatever is around you, the calibration is still needed to negate the Trackmo, but now you may also have negated the external interference.
I think Eric is right - if I calibrate with the Trackimo on, then the AC will compensate for whatever interference it is causing and will carry that compensation with it throughout the flight.
I don't think that is what I was trying to say, though one might conclude that.
Rather, regardless of what one has on their AC, if the RC doesn't agree with the direction the nose of the drone is pointing, then there is magnetic interference with the AC's compass....time to find a new takeoff spot.
If you calibrate the compass with the Trackimo on it, whatever interference it is causing will be taken into account in the calibration and will be corrected by that calibration. So, once calibrated with the Trackimo attached, it should show true north when the aircraft is pointing north. Right? Or am I missing something?
I am not an expert, but logically as long as the AC & RC agree on the compass heading, then the flight should be good to go.
From what I have read in other threads, it is when the AC's compass disagrees with the RC (aka the DJI Go app) that the control systems get into a disagreement with the possibility the AC behaves erratically.
Hopefully the experts can weigh in here, or you might search the forums for IMU and FC disagreement scenario.
From what I have seen, it is when the compass disagrees wit the IMU that you have problems. That is making re-think calibrating it with the Trackimo on it and thinking it will be okay. I am not sure where the GoApp gets it heading info from. If it is from the IMU and I have calibrated with the Trackimo on and the app looks like the AC direction is correct that would be fine. But if the app is getting the info from the compass only, it still might be in disagreement with the IMU, though I would think that would give compass or IMU error warnings even while it is still on the ground. I just don't know. Maybe one of thee very smart people on this thread can weigh in.
After some further thought on the matter, I am not sure whether or not calibration with the Trackimo mounted is the answer. Any magnetic interference from the Trackimo's electronics will indeed be constant with respect to the AC, but it will not be constant with respect to magnetic north.
Oversimplifying, and ignoring that there are two compasses on the aircraft, assume for example that the AC is pointed at due magnetic north and the Trackimo is creating even a weak magnetic field that "attracts" the AC and the Trackimo is mounted at the rear of the aircraft. If the AC is pointed due north, the "pull" of the Trackimo will be toward the south and therefore on the north-south magnetic line and will not affect the AC's reading of magnetic north. The same goes for the situation where the AC is pointed due south. However, when the aircraft is pointed some other direction, there is going to be an east or west error. Assume in our situation that the AC is pointed east but the Trackimo is "pulling" it west, the AC compasses will read that the AC is pointed east of north. Or, if pointed west, the AC compass will read it as west of magnetic north..
Never understood why before, but when I used to fly single engine airplanes, the compass correction card always mounted in them, would give different magnetic corrections for different compass readings. North might be one degree of correction while east might be 3 degrees correction, south 1 degree and west 2 degrees.
I am guessing that the magnetic field form the Trackimo is fairly weak, though I would also guess that the compasses in the Mavic are fairly sensitive. On the last set of test flights where I was getting no errors without the Trackimo mounted but compass errors with it mounted, according to Airdata, those compass errors were low risk. So maybe it is still safe to calibrate without the Trackimo on the AC and maybe any compass error induced is small enough not to cause serious issues. I have heard that if the IMU heading and the compass heading disagree there is likely to be a fly-away, but maybe in my case, the disagreement is small enough not to matter much. More testing is in order. Or at least so I hope.
I would also note that I experienced problems calibrating the compass with the Trackimo mounted. It simply would not calibrate at times,
I hope someone more knowledgeable than I see this thread and comments.
Oversimplifying, and ignoring that there are two compasses on the aircraft, assume for example that the AC is pointed at due magnetic north and the Trackimo is creating even a weak magnetic field that "attracts" the AC and the Trackimo is mounted at the rear of the aircraft. If the AC is pointed due north, the "pull" of the Trackimo will be toward the south and therefore on the north-south magnetic line and will not affect the AC's reading of magnetic north. The same goes for the situation where the AC is pointed due south. However, when the aircraft is pointed some other direction, there is going to be an east or west error. Assume in our situation that the AC is pointed east but the Trackimo is "pulling" it west, the AC compasses will read that the AC is pointed east of north. Or, if pointed west, the AC compass will read it as west of magnetic north..
Never understood why before, but when I used to fly single engine airplanes, the compass correction card always mounted in them, would give different magnetic corrections for different compass readings. North might be one degree of correction while east might be 3 degrees correction, south 1 degree and west 2 degrees.
I am guessing that the magnetic field form the Trackimo is fairly weak, though I would also guess that the compasses in the Mavic are fairly sensitive. On the last set of test flights where I was getting no errors without the Trackimo mounted but compass errors with it mounted, according to Airdata, those compass errors were low risk. So maybe it is still safe to calibrate without the Trackimo on the AC and maybe any compass error induced is small enough not to cause serious issues. I have heard that if the IMU heading and the compass heading disagree there is likely to be a fly-away, but maybe in my case, the disagreement is small enough not to matter much. More testing is in order. Or at least so I hope.
I would also note that I experienced problems calibrating the compass with the Trackimo mounted. It simply would not calibrate at times,
I hope someone more knowledgeable than I see this thread and comments.
Last edited:
The compass(es) are magnetic flux detectors. There are components on the AC, the motors for example, that the magnetic sensors will pick up. That's negated when you do a calibration. As you rotate the AC, the magnetic field rises and falls in relation to external magnetic fields, such as earth, but the magnetic influence on the AC is sensed the same. The bias magnetic field is subtracted out.
The RC heading is obtained by your device compass if it has one. Much of the time it can be out of calibration, especially with Androids. Because of the device compass issue, the direction the app claims you are facing might not be which way you are actually facing. The AC compass is more likely to be accurate.
So if both you and the AC are facing the same way, just because the map shows facing different directions doesn't necessarily mean the AC is experiencing magnetic interference.
However in the map view you can set it to North Up or just go by the landmarks shown on the map as an accurate heading reference. If you're not sure which way is north and you're unfamiliar with the area landmarks, you don't have any accurate means of verifying AC compass calibration and external interference.
The RC heading is obtained by your device compass if it has one. Much of the time it can be out of calibration, especially with Androids. Because of the device compass issue, the direction the app claims you are facing might not be which way you are actually facing. The AC compass is more likely to be accurate.
So if both you and the AC are facing the same way, just because the map shows facing different directions doesn't necessarily mean the AC is experiencing magnetic interference.
However in the map view you can set it to North Up or just go by the landmarks shown on the map as an accurate heading reference. If you're not sure which way is north and you're unfamiliar with the area landmarks, you don't have any accurate means of verifying AC compass calibration and external interference.
The Go App heading indicator, the red arrow on the map display, is a reflection of what the Flight Controller thinks the Yaw value is. The FC uses a complicated algorithm to determine it's Yaw value. At batteryOn Yaw is set to the compass value. After that it's mostly the rate gyros that are used with the compass being used at a low gain to correct the errors seen in the gyros. It is usually the case that the geomagnetic distortion causing a compromised compass will have limited extent - less than a meter. As the AC leaves this distortion the compass value will become correct but the Yaw will still be incorrect since there has been no actual rotation that would be seen by the rate gyros.
The geomagnetic declination is the difference between the magnetic heading and the true heading. Where I live magnetic the north pole is 13° to the right of the earth's north pole - where all the longitude lines meet. So the declination here is 13° and the FC needs to know that in order to navigate properly. Declination can not be determined by the compass calibration process. Rather declination is determined by knowing the Lat/Long coordinates that is then used by an algorithm that computes the declination for that location.
To repeat, a compass calibration can not determine, or compensate for, declination. The same is true for any local magnetic effects that cause the compass to have an incorrect heading. These effects are external to the AC and the calibration process can not compensate for them.
However, it is possible for a compass calibration to determine the geomagnetic field strength. Knowing this may be helpful in determining if the AC is sitting in a geomagnetic distortion. Undistorted, the geomagnetic field has uniform strength. A distortion will result in a non-uniform strength. Some points will be lower and some higher. A field strength is outside of the normal range may be due to a distorted field. The "magnetic interference" display in the Go App is just a measure of much the field strength deviates rom the acceptable range.
The geomagnetic declination is the difference between the magnetic heading and the true heading. Where I live magnetic the north pole is 13° to the right of the earth's north pole - where all the longitude lines meet. So the declination here is 13° and the FC needs to know that in order to navigate properly. Declination can not be determined by the compass calibration process. Rather declination is determined by knowing the Lat/Long coordinates that is then used by an algorithm that computes the declination for that location.
To repeat, a compass calibration can not determine, or compensate for, declination. The same is true for any local magnetic effects that cause the compass to have an incorrect heading. These effects are external to the AC and the calibration process can not compensate for them.
However, it is possible for a compass calibration to determine the geomagnetic field strength. Knowing this may be helpful in determining if the AC is sitting in a geomagnetic distortion. Undistorted, the geomagnetic field has uniform strength. A distortion will result in a non-uniform strength. Some points will be lower and some higher. A field strength is outside of the normal range may be due to a distorted field. The "magnetic interference" display in the Go App is just a measure of much the field strength deviates rom the acceptable range.
Both comments are very helpful. I will check the magnetic interference display in the Go App when I test fly it next. Thanks
This is a link to the thread I was trying to remember....Sar104 did a great job in explaining how compass errors occur and the disastrous results that may follow. I cannot say I understood the post in all its detail, but I did appreciate how the FC/IMU/compass interact. Have a look at it.
A short explanation of compass function, calibration and errors.
BACKGROUND The aircraft compass is often misunderstood, both in terms of function and purpose, but it is important to understand what the compass measures, why, and what happens to that information. At the most basic level the purpose of the compass is to measure the earth’s magnetic field in...
mavicpilots.com
I am making progress, and may have found the source of the problem. Thanks to SAR104's explanation of compass, calibration and errors, I realized that in the advanced setting not only is IMU health shown but you can toggle to compass and see magnetic interference for each compass. Not in the Mavic Pro manual that I remember.
So I first did a compass calibration in an area with no compass interference. I made a cardboard compass rose and laid the ac on top and checked the Go App in each of the four cardinal directions and it was pointing correctly in the app.
Next I went to the Advanced settings in the GO4 App and checked for magnetic interference, first picture. Was little interference shown for compass 1 and slightly more for compass 2. Both were green, which means excellent per the App.
Next I put on the Trackimo in its rear mount. I checked for interference again. Picture 2 and found more for compass 1 and about the same for compass 2. Both green.
Next I put on the Tile which I had mounted with Velcro on top of the battery. Magnetic interference went nuts. Picture 3.
Next I took off the Tile and Trackimo and mounted the strobe by Velcro at the back of the body. Slightly more interference than with nothing mounted. Picture 4
Next I left the strobe and mounted the Trackimo. Sometimes it showed green Excellent and sometimes at the very low end of the yellow good. Picture 5
The one thing I neglected to do was check the interference with Trackimo and strobe mounted in each of the cardinal directions but I suspect it will show as excellent or good.
At least as of now, subject to change, it appears that the culprit is the Tile. Kind of a shame because you can use its Bluetooth to locate the Tile if nearby and also make it sound beeps so that you can hear it if close enough. Would be a nice aid to locating it if it goes down in brush. But I will live without it.
I will start a separate thread about the Tile causing problems and suggest people who want to try to use it test for magnetic interference in the app.
Thanks to all who helped me out in this thread and to SAR104 for his excellent tutorial on the compass and Eric for pointing me to it.
If anyone sees any errors or dangerous assumptions in this post please let me know.
And again, thanks to all,
So I first did a compass calibration in an area with no compass interference. I made a cardboard compass rose and laid the ac on top and checked the Go App in each of the four cardinal directions and it was pointing correctly in the app.
Next I went to the Advanced settings in the GO4 App and checked for magnetic interference, first picture. Was little interference shown for compass 1 and slightly more for compass 2. Both were green, which means excellent per the App.
Next I put on the Trackimo in its rear mount. I checked for interference again. Picture 2 and found more for compass 1 and about the same for compass 2. Both green.
Next I put on the Tile which I had mounted with Velcro on top of the battery. Magnetic interference went nuts. Picture 3.
Next I took off the Tile and Trackimo and mounted the strobe by Velcro at the back of the body. Slightly more interference than with nothing mounted. Picture 4
Next I left the strobe and mounted the Trackimo. Sometimes it showed green Excellent and sometimes at the very low end of the yellow good. Picture 5
The one thing I neglected to do was check the interference with Trackimo and strobe mounted in each of the cardinal directions but I suspect it will show as excellent or good.
At least as of now, subject to change, it appears that the culprit is the Tile. Kind of a shame because you can use its Bluetooth to locate the Tile if nearby and also make it sound beeps so that you can hear it if close enough. Would be a nice aid to locating it if it goes down in brush. But I will live without it.
I will start a separate thread about the Tile causing problems and suggest people who want to try to use it test for magnetic interference in the app.
Thanks to all who helped me out in this thread and to SAR104 for his excellent tutorial on the compass and Eric for pointing me to it.
If anyone sees any errors or dangerous assumptions in this post please let me know.
And again, thanks to all,
Attachments
Similar threads
DJI Drone Deals
New Threads
-
-
-
Air 3s Video settings tested - D-Log vs. HLG- Started by DroneZone Samui
- Replies: 1
-
-
Mavic 3 classic purchased in the USA, on CE transmission profile?
- Started by Ossbourne
- Replies: 1
Forum statistics
