As stated previously, the Phanrom's offsets from the CG are fixed out of the box, so no offsets are needed. In the Flamwheel, or for those who replace the standard compass with the combination GPS/compass that normally comes with the Flamewheels, the compass is part of the GPS, and is included when you measure the Z distance to the CG. Go ahead and put your compass on a pylon and report back.
phantom suddenly possessed in GPS mode
- Thread starter MILLER4PRESIDENT2020
- Start date
GeneL said:
Sorry mate, I have to call BS on that. If this were the case, it would be documented somewhere, don't you think?
GeneL said:
Okay, I moved my compass up the leg 1/2". It made no difference in the flight characteristics. Happy?
I would caution you guys to not believe everything that you hear.
Dave Pitman said:
http://www.youtube.com/watch?v=3joOzwHa-Uw
Compass module is very sensitive to electromagnetic field as well as the simple Ferrous metals like metal bars underneath the concrete pavement. So this is the first reason why the compass has to be located not only afar but specifically spot where the interference is to be the least.
Next, as per above link the compass module does not only report the X,Y coordinate but Z value too. To complicate the matter even more, our Phantom does "tilt" in flight so the tilt compensation will be done by the accelerometer. So this is the second reason why the compass has to be at the predetermined position or should I say relative vertical distance with respect to the NAZA.
Phantom compass is separated from the GPS and its offset position can NOT be programmed like the GPS, thus keep it stock.
Sorry that I can NOT provide you the specific documents about this compass relative vertical distance but I tried my best to explain as above.
BTW, Phantom document never mentions that you shouldn't take off your Phantom from the car roof either. I tried that and it messed my compass! However, there is a warning about flying nearby the ferrous metal.
I watched the video. I don't really get how it relates to this discussion, other than the importance of keeping the compass away from magnetic fields; No argument with that.
I don't see any evidence that the compass would be compromised by a change in vertical height. Perhaps you can point to what time in the video that is documented. As I stated above, I moved my compass 1/2" vertically from stock, and there was no noticeable difference.
Also, the Phantom documentation does say to keep the compass away from things like speakers which have a magnetic field. And, to not calibrate the compass near magnetic fields or interference.
I don't see any evidence that the compass would be compromised by a change in vertical height. Perhaps you can point to what time in the video that is documented. As I stated above, I moved my compass 1/2" vertically from stock, and there was no noticeable difference.
Also, the Phantom documentation does say to keep the compass away from things like speakers which have a magnetic field. And, to not calibrate the compass near magnetic fields or interference.
tanasit said:http://www.youtube.com/watch?v=3joOzwHa-Uw
Compass module is very sensitive to electromagnetic field as well as the simple Ferrous metals like metal bars underneath the concrete pavement. So this is the first reason why the compass has to be located not only afar but specifically spot where the interference is to be the least.
Next, as per above link the compass module does not only report the X,Y coordinate but Z value too. To complicate the matter even more, our Phantom does "tilt" in flight so the tilt compensation will be done by the accelerometer. So this is the second reason why the compass has to be at the predetermined position or should I say relative vertical distance with respect to the NAZA.
Phantom compass is separated from the GPS and its offset position can NOT be programmed like the GPS, thus keep it stock.
Sorry that I can NOT provide you the specific documents about this compass relative vertical distance but I tried my best to explain as above.
BTW, Phantom document never mentions that you shouldn't take off your Phantom from the car roof either. I tried that and it messed my compass! However, there is a warning about flying nearby the ferrous metal.
Dave Pitman said:
Ref-1:: page 10 Quick Start Manual V1.7;
DO NOT use the magnetic screwdriver. Otherwise, keep the screwdriver at least 10cm away from the compass
module, to avoid magnetic interference.
Ref-2: page 11 Quick Start Manula V1.7;
After the system start and self-checking has finished, if the LED blinks Red, Yellow and Green continually, that
means the IMU data is abnormal. The PHANTOM will not work, please connect to the Assistant Software and
follow the tips to do operation. If it blinks red and yellow lights alternately ( ), that means the
compass error is too big, it can be caused by the following three cases. Please connect to the Assistant Software,
select the “tools” tab and follow the tips of the “IMU Calibration” to do operation.
1. There are ferromagnetic substance around; first make sure that the compass has been calibrated correctly, you
can lift the aircraft up (about 1m from the ground), and stay away from the surrounding possible ferromagnetic
material object, if there is no red and yellow flashing after lifting it up about 1m from the ground, then it will not
affect the flight.
2. The compass module had been put near a magnet; in this situation please timely replace the compass for a new
one, otherwise it will lead to some abnormal action, or even loss of control.
3. The compass is not properly calibrated; in this situation please calibrate the compass correctly again, please see
the GPS compass calibration for details.
Ref-3: page 13;
Make sure to keep away from ferromagnetic substance and other electronic equipment when calibrating or flying. If you keep having calibration
failure, it might suggest that there is magnetic interference or other ferromagnetic substance, please avoid flying in this area.
That being said, as for the vertical distance (Z value):
"....compass module does not only report the X,Y coordinate but Z value too. To complicate the matter even more, our Phantom does "tilt" in flight so the tilt compensation will be done by the accelerometer."
The compass Z value must be taken into account in order for the NAZA to operate properly as well as the Z valve from the GPS receiver. Vertical input may have less effect than the X and Y, thus the ill effect may not be obvious if hovering. Non vertical compass has been known to cause the irregular in flying circle and the improper MC orientation and location will at least cause the straight line flight off the home point location.
DJI does NOT provide the documents but suggests to keep the compass location as per stock. So if one opts to choose different locations, one shouldn't expect the same performance without testing to find out the exact possible "off position" from the manufacture specification. The suggestion from the DJI never stops anyone to try something else anyway. This is how and why we have this wonderful forum.
RE: I don't see any evidence that the compass would be compromised by a change in vertical height.
May I suggest you mount your compass with extreme vertical distance (much more than 1/2" that you already tried) and see if you can see the evidence and kindly report back.
I believe when first the Phantom came out, DJI did not expect numerous after market parts, in particular the landing gears which directly involve the compass installation. In addition, there is no provision for setting the X,Y,Z values for the compass which is highly unlikely that one will relocate the compass, unlike the GPS. Thus DJI choose not to provide documents regard the compass orientation and location details but keep a simple suggestion to keep it as stock.
May I suggest you mount your compass with extreme vertical distance (much more than 1/2" that you already tried) and see if you can see the evidence and kindly report back.
I believe when first the Phantom came out, DJI did not expect numerous after market parts, in particular the landing gears which directly involve the compass installation. In addition, there is no provision for setting the X,Y,Z values for the compass which is highly unlikely that one will relocate the compass, unlike the GPS. Thus DJI choose not to provide documents regard the compass orientation and location details but keep a simple suggestion to keep it as stock.
Hey tanasit,
While I agree with everything you highlighted in red out of the manual, you still give no data to go with your claim that the vertical offset distance of the compass is important. You mention that "DJI does NOT provide the documents but suggests to keep the compass location as per stock." Please provide a link to this suggestion by dji.
Think about this. When you hook up your Phantom to the Naza assistant application, the application does not know the difference whether the Naza is in a Phantom, or in a F550, or in some other model completely. So, if the compass is in a puck instead of mounted on the leg of the Phantom, it will be in a very different vertical position, but the assistant application does not know the difference, and does not care. It only cares about the offset of the gps, and the orientation of the compass in regard the the center line of the naza, not the vertical offset of the sensor itself. When using a puck, the compass will be oriented properly if the gps is oriented properly, the two are fixed. On the Phantom, the owner must make sure that the compass sensor is oriented properly, especially if using different landing gear. But the vertical offset makes no difference.
You don't have to believe me, it doesn't matter. Just don't make the claim that the vertical offset is important when it is clearly not, and you have nothing to back up that claim. I am not angry, and I hope you are not either. We can let this debate end, and we can each believe what we want. I did move my compass as much as I could without the orientation being changed, and had no problems, but you discount this evidence. I have no way or interest in trying to figure out how to mount it a great distance different just to prove a point.
Cheers.
While I agree with everything you highlighted in red out of the manual, you still give no data to go with your claim that the vertical offset distance of the compass is important. You mention that "DJI does NOT provide the documents but suggests to keep the compass location as per stock." Please provide a link to this suggestion by dji.
Think about this. When you hook up your Phantom to the Naza assistant application, the application does not know the difference whether the Naza is in a Phantom, or in a F550, or in some other model completely. So, if the compass is in a puck instead of mounted on the leg of the Phantom, it will be in a very different vertical position, but the assistant application does not know the difference, and does not care. It only cares about the offset of the gps, and the orientation of the compass in regard the the center line of the naza, not the vertical offset of the sensor itself. When using a puck, the compass will be oriented properly if the gps is oriented properly, the two are fixed. On the Phantom, the owner must make sure that the compass sensor is oriented properly, especially if using different landing gear. But the vertical offset makes no difference.
You don't have to believe me, it doesn't matter. Just don't make the claim that the vertical offset is important when it is clearly not, and you have nothing to back up that claim. I am not angry, and I hope you are not either. We can let this debate end, and we can each believe what we want. I did move my compass as much as I could without the orientation being changed, and had no problems, but you discount this evidence. I have no way or interest in trying to figure out how to mount it a great distance different just to prove a point.
Cheers.
tanasit said:
@dave pitman, the assistant software can tell between a phantom naza and a standard naza v1 or v2 or lite!
The compass settings are set in the firmware, use a v2 naza and you'll need to change to v2 GPS puck with internal compass, because the v2 firmware won't accept a phantom firmware!
I've just extended my landing gear (for a tarot gimbal) my compass is about 15mm lower than it was, I was stressing that it will have issues, I haven't moved it because I only have magnetic screwdrivers that small
As pointed out earlier, the compass sense's the earths magnetic field, this is where it gets tricky! The magnetic field data is basically a 2 axis graph (unless your looking at the "whole" field from "outside" the mag field, while we are on earth, we don't need the 3rd axis), think of the display on a handheld compass is a 2d graph! For eg, Due east will send you around the earth with no change in the y value (if earth is layed out as a 2d map).
What make the phantom compass so tricky is that the phantom tilts, so it needs to pull specific 2d data from a sensor being subjected to 3d movements! To develop the algorithm to deal with the 3rd axis, you would need to know the exact position of the sensor from the 3 dimensional centreline, not the COG, but effectively the pivot and tilt point, funnily enough the CoP/T is almost exactly where the GPS puck is!
When any other version of naza is used, the X,Y,Z position (in relation to the naza) of the GPS is input in the software, this also gives the compass location data for the algorithm to convert the 3d to 2d data!
The compass settings are set in the firmware, use a v2 naza and you'll need to change to v2 GPS puck with internal compass, because the v2 firmware won't accept a phantom firmware!
I've just extended my landing gear (for a tarot gimbal) my compass is about 15mm lower than it was, I was stressing that it will have issues, I haven't moved it because I only have magnetic screwdrivers that small
As pointed out earlier, the compass sense's the earths magnetic field, this is where it gets tricky! The magnetic field data is basically a 2 axis graph (unless your looking at the "whole" field from "outside" the mag field, while we are on earth, we don't need the 3rd axis), think of the display on a handheld compass is a 2d graph! For eg, Due east will send you around the earth with no change in the y value (if earth is layed out as a 2d map).
What make the phantom compass so tricky is that the phantom tilts, so it needs to pull specific 2d data from a sensor being subjected to 3d movements! To develop the algorithm to deal with the 3rd axis, you would need to know the exact position of the sensor from the 3 dimensional centreline, not the COG, but effectively the pivot and tilt point, funnily enough the CoP/T is almost exactly where the GPS puck is!
When any other version of naza is used, the X,Y,Z position (in relation to the naza) of the GPS is input in the software, this also gives the compass location data for the algorithm to convert the 3d to 2d data!
i just tore apart my phantom.1.1.1 and plan on moving everything over to a tbs discovery.
ive already gotten the motors and speed controls in as well as the naza and core.
now im stumped as to where to go. my naza has the external compass and the puck has no housing
ive made a plastic housing for the gps and ordered a pole mount, but now I need to attach the compass and don't know where to put it.
any help would be greatly appreciated. another thing ive ran into is how to power the naza? the phantom had a separate board. will I need a naza pmu also?
ive already gotten the motors and speed controls in as well as the naza and core.
now im stumped as to where to go. my naza has the external compass and the puck has no housing
ive made a plastic housing for the gps and ordered a pole mount, but now I need to attach the compass and don't know where to put it.
any help would be greatly appreciated. another thing ive ran into is how to power the naza? the phantom had a separate board. will I need a naza pmu also?
