I know this has probably been gone over a thousand times but I would just like to try and get some official word on DJI drone calibrations if possible.  Hopefully someone nice and helpful from DJI will have an answer to some of the below for me and others


There are hundreds of differing opinions around the web in regards to compass and IMU calibrations ranging from "It's pointless because it doesn't do anything" through to "I do it every time I start the drone or there will be hellfire and damnation rained down on the planet"

Apart from the obvious times that the equipment tells you to do it I am of the opinion that recalibrating a calibrated device every now and then can not be a bad thing on a portable piece of equipment that gets bumped around and can be affected by outside influences, So my questions are:

What sensors do the calibrations check?
Does calibration actually achieve anything?
When should it be done?
How often should it be done?

Apart from the obvious times that the equipment tells you to do it I am of the opinion that recalibrating a calibrated device every now and then can not be a bad thing on a portable piece of equipment that gets bumped around and can be affected by outside influences,

A better idea is to learn what compass calibration actually does - it's not what many people imagine.

Labroides Posted at 4-12 22:08
Apart from the obvious times that the equipment tells you to do it I am of the opinion that recalibrating a calibrated device every now and then can not be a bad thing on a portable piece of equipment that gets bumped around and can be affected by outside influences,

A better idea is to learn what compass calibration actually does - it's not what many people imagine.

Thanks Labroides.  I agree and I get that but it is part of the reason for asking the questions.  I am hoping DJI will clarify and that this will be somewhere people can refer to their official response and use it as a knowledge tool for why and what all of the calibrations do.  The questions do also refer to the IMU though not just the compass.

Calibrating the compass and IMU (Inertial Measurement Unit) on a drone helps to ensure that the drone's flight controls are accurate and stable. Here's what each calibration does:

Compass Calibration:
The compass on a drone is used to determine the drone's orientation and heading. If the compass is not calibrated correctly, the drone's flight controls may become erratic or unstable. Compass calibration involves rotating the drone around all three axes (yaw, pitch, and roll) while the drone's GPS and compass sensors detect the magnetic field. This allows the drone's flight controller to accurately determine the drone's orientation and heading.

IMU Calibration:
The IMU on a drone is a set of sensors that measure the drone's linear and angular acceleration, as well as its rotational rate. IMU calibration involves placing the drone on a level surface and allowing the sensors to detect the gravity and other forces acting on the drone. This allows the drone's flight controller to accurately measure the drone's position and movement.

Calibrating the compass and IMU is important for ensuring the accuracy and stability of the drone's flight controls. A properly calibrated drone will be able to fly & hover more smoothly and accurately, reducing the risk of crashes or other accidents. It is recommended to calibrate the compass and IMU on a regular basis, especially if you have traveled to a new location or if you notice any issues with the drone's flight performance.

I usually check its hovering behavior & if its drifting I calibrate it.

djiuser_eFdo5aGvsfJ3 Posted at 4-12 22:37
Thanks Labroides.  I agree and I get that but it is part of the reason for asking the questions.  I am hoping DJI will clarify and that this will be somewhere people can refer to their official response and use it as a knowledge tool for why and what all of the calibrations do.  The questions do also refer to the IMU though not just the compass.

I am hoping DJI will clarify and that this will be somewhere people can refer to their official response and use it as a knowledge tool for why and what all of the calibrations do.
You're in the wrong place for that.
The moderators don't fly or have technical knowledge.
btw ... the time when compass calibration would be advised is when you modify or rebuild the drone or add accessories to it.

The questions do also refer to the IMU though not just the compass.
The IMU is fine without recalibrating it.
If it was out of calibration, you would know.
The only time I'd recalibrate the IMU would be after a crash or serious impact on the drone, otherwise you can leave it untouched for years, just like the compass.

Ranjan Posted at 4-12 23:22
Calibrating the compass and IMU (Inertial Measurement Unit) on a drone helps to ensure that the drone's flight controls are accurate and stable. Here's what each calibration does:

Compass Calibration:

It is recommended to calibrate the compass and IMU on a regular basis, especially if you have traveled to a new location or if you notice any issues with the drone's flight performance.
This is the myth.
You won't find any such recommendation in the manual for your drone.

The only people who make that recommendation are people who don't know what those calibrations actually do.
The compass calibration does not change, no matter how far away your flying site is.
It has nothing to do with where you are on earth.

Remembering magnetic declination
Killed in Action on April 13, 2023 by a fellow DJI forum member.
Rest in peace, little angel gone too soon.

LV_Forestry Posted at 4-13 02:07
Remembering magnetic declination
Killed in Action on April 13, 2023 by a fellow DJI forum member.
Rest in peace, little angel gone too soon.

I'm not sure what your point is?

But if you think that magnetic declination is somehow related to compass calibration, that myth was disproved years ago.

As Labroides wrote, the IMU does not frequently require calibration. The most frequent case is during large temperature changes because these components are sensitive to them. Anyway, don't panic, if necessary, the application will tell you and ask you to put the drone in different positions in order to perform the automatic calibration.

For the compass it's the same, the North South axis is far from perfect. It looks like this:

source : Magnetic declination - Wikipedia

Does the compass needle point toward the magnetic pole?
No. The compass points in the directions of the horizontal component of the magnetic field where the compass is located, and not to any single point.

So when you move a certain distance, the magnetic declination will change. Again the application will ask you to initiate the calibration step to perform the correction. Same with big temperature changes.


If you want more info you can go see the compass datasheets. You will see that there is a self-calibration process. It often requires moving the sensor in more than one direction. This is the reason why the application asks you to rotate the drone on Z with two different angles. For most mobile phones equipped with a compass, you have to make "eight" patterns.

I put the link to the datasheet of a Honeywell compass, I don't know if it's this model that is used in DJI drones. Simply that I worked with this reference so I know a little about what it is made of. And yes it requires regular calibration, you can look at the self test chapter it is well explained.
Datasheet - honeywellaerospace-hmc5883ltr-datasheets-7405.pdf (jakelectronics.com)

To sum up:

If the application asks you to calibrate the compass it is quite normal if you have moved a certain distance, in my experience about 50km but it all depends on where you are on earth. Refer to the map above.
If the application asks you for a calibration each time the drone is started, this is not normal. It is then necessary to make sure that there is not a metallic mass nearby, a car for example.

For the IMU, it must be really exceptional (big temperature change). If the application asks for calibration every time, think about to send it to the service.

Labroides Posted at 4-13 02:16
I'm not sure what your point is?

But if you think that magnetic declination is somehow related to compass calibration, that myth was disproved years ago.

I don't know who disapproved of it, but it wasn't me, I assure you.

The relationship between Geographic coordinates and compass is heading.

Imagine that you fly the drone over the equator. The latitude will remain at zero, the longitude will vary.

So the heading is East 90 or West 270 depending on which way you are moving.

If the magnetic declination is not correctly established, there will be a mismatch between compass heading and GNSS heading. The EKF will not appreciate at all. It's as simple as that.

Look at an indisputable example, it is aeronautical documentation. The distance between the two VOR shown on the map is 44km.
One is 7 degrees East, and the other 10 degrees East. That's huge for such a small gap on the scale of the planet.

LV_Forestry Posted at 4-13 02:07
Remembering magnetic declination
Killed in Action on April 13, 2023 by a fellow DJI forum member.
Rest in peace, little angel gone too soon.

That was beautiful...

Thank you

LV_Forestry Posted at 4-13 02:25
As Labroides wrote, the IMU does not frequently require calibration. The most frequent case is during large temperature changes because these components are sensitive to them. Anyway, don't panic, if necessary, the application will tell you and ask you to put the drone in different positions in order to perform the automatic calibration.

For the compass it's the same, the North South axis is far from perfect. It looks like this:

I thought this myth had died away, but it seems there are still some hangers on.
If the application asks you to calibrate the compass it is quite normal if you have moved a certain distance, in my experience about 50km but it all depends on where you are on earth.

You don't have to tell me what magnetic declination is, I've sailed a yacht over 12000 miles so I'm well aware of it.
But none of that has any relation to drone flying or your drone's compass.
Compass calibration has nothing at all to do with where you are or how far you are from where you last flew.
Knowing how your drone works makes you a much better flyer than someone who relies on myths.

ps ... DJI contributed to the myths around compass calibration with the misinformation they used to put in their manuals for years.
If you had a Mavic 2, DJI for a reason known only to them, they would force the operator to (completely unnecessarily) recalibrate the compass if you moved some distance (50 km?) or it was >30 days since your last compass calibration.
There was no physical reason for this.
Fortunately their manuals have now dropped the compass misinformation, but only after spreading myths for a long time.

Labroides Posted at 4-13 02:52
I thought this myth had died away, but it seems there are still some hangers on.
If the application asks you to calibrate the compass it is quite normal if you have moved a certain distance, in my experience about 50km but it all depends on where you are on earth.

And I am proud to announce that on the M300 there is no longer any need to perform compass calibration.

But on M2, P4, Pixhawk, Ardupilot, it's always the same. Calibration of the compass should be done regularly. Magnetic declination or not, the temperature has an extremely important role which justifies the calibration on its own.

LV_Forestry Posted at 4-13 02:59
And I am proud to announce that on the M300 there is no longer any need to perform compass calibration.

But on M2, P4, Pixhawk, Ardupilot, it's always the same. Calibration of the compass should be done regularly. Magnetic declination or not, the temperature has an extremely important role which justifies the calibration on its own.

But on M2, P4, Pixhawk, Ardupilot, it's always the same. Calibration of the compass should be done regularly. Magnetic declination or not, the temperature has an extremely important role which justifies the calibration on its own
There neever was any need to recalibrate the compass of the M300.
I don't know about Pixhawk or Ardupilot, but if you are suggesting that there's any need to recalibrate the compass of the Phantom 4, that's 100% untrue.
If DJI firmware still enforces it for the Mavic 2, there is no physical reason for it.

Labroides Posted at 4-13 03:03
But on M2, P4, Pixhawk, Ardupilot, it's always the same. Calibration of the compass should be done regularly. Magnetic declination or not, the temperature has an extremely important role which justifies the calibration on its own
There neever was any need to recalibrate the compass of the M300.
I don't know about Pixhawk or Ardupilot, but if you are suggesting that there's any need to recalibrate the compass of the Phantom 4, that's 100% untrue.

"There neever was any need to recalibrate the compass of the M300"


Yes iam not even sure that there is one. It uses his dual GNSS antenna to get azimuth

Burt37 Posted at 4-13 02:47
That was beautiful...

Thank you

LV_Forestry Posted at 4-13 03:10
"There neever was any need to recalibrate the compass of the M300"

I've never flown an M300, but I'm pretty sure that it has a compass.
Otherwise this video makes no sense.

Another thing I suspect makes no sense is the idea that any number of GNSS receivers could provide azimuth data.

But what the M300 has or hasn't isn't really relevant to the topic of this thread.

Burt37 Posted at 4-13 02:47
That was beautiful...

Thank you

That was beautiful...
But it's completely irrelevant to flying your DJI drone.

Labroides Posted at 4-13 03:21
I've never flown an M300, but I'm pretty sure that it has a compass.
Otherwise this video makes no sense.
https://www.youtube.com/watch?v=eKnAVgkQtUs

"Another thing I suspect makes no sense is the idea that any number of GNSS receivers could provide azimuth data"


As much on the initial subject of the thread I am open to discussion, you may be right about the magnetic declination because after all why not.


But on the fact that it is possible to have the azimuth with at least two GNSS receivers, I am 100% certain. On forestry machines, I need the heading to calculate trajectories and above all to know the direction in which the machine is moving. At first I used a compass, but there was too much of a problem with deviation when the tractor changed location, and especially when the temperature dropped or rose sharply, typical in the fall and spring. I switched to a GNSS-only solution, placing two antennas with two ZF9P receivers. One is master and compares its position with the slave, the exact dimension between the two is known, so we can do local RTCM and therefore know precisely the azimuth without outside help.


simpleRTK2B+heading hookup guide - ArduSimple

Pure GNSS, no compass needed.

LV_Forestry Posted at 4-13 04:03
"Another thing I suspect makes no sense is the idea that any number of GNSS receivers could provide azimuth data"

I've not used an M300 or heard of using 2x GNSS to somehow get azimuth data and can't imagine how it could.
But maybe the M300 does all sorts of magic tricks.
However that has nothing to do with how our Mavics work and what you need to know abouyt the compass, a subject that is surrounded by much myth and misunderstanding (as demonstrated earlier in this thread.
DJI themselves contributed to the misunderstanding by putting misinformation in their manuals for years.

To find out the facts, the best resource is the first post in this much too lengthy thread from another forum.
https://mavicpilots.com/threads/ ... n-and-errors.90792/

Labroides Posted at 4-13 04:11
I've not used an M300 or heard of using 2x GNSS to somehow get azimuth data and can't imagine how it could.
But maybe the M300 does all sorts of magic tricks.
However that has nothing to do with how our Mavics work and what you need to know abouyt the compass, a subject that is surrounded by much myth and misunderstanding (as demonstrated earlier in this thread.

Well, in the thread you mention everything is said.
By moving the drone the calibration process can separate the terrestrial magnetic field from that of the drone.
So this operation must be performed if the magnetic field of the drone or the earth's magnetic field (magnetic declination) changes.
Then the EKF has the right info to provide the true heading.
This is exactly what I observe with the assembly done on forestry machines. I agree !


At the end of the post he changes his mind and says it has no relation. While he writes it just above.
What DJI engineers have been able to change is the way the EKF processes information. Very likely.

LV_Forestry Posted at 4-13 04:33
Well, in the thread you mention everything is said.
By moving the drone the calibration process can separate the terrestrial magnetic field from that of the drone.
So this operation must be performed if the magnetic field of the drone or the earth's magnetic field (magnetic declination) changes.

By moving the drone the calibration process can separate the terrestrial magnetic field from that of the drone.
So this operation must be performed if the magnetic field of the drone or the earth's magnetic field (magnetic declination) changes.
No .. the calibration process identifies and measures the magnetic fields that are part of the drone as opposed to those that are outside of the drone (no matter what or where they are)
Magnetic declination has nothing to do with it.

Labroides Posted at 4-13 04:58
By moving the drone the calibration process can separate the terrestrial magnetic field from that of the drone.
So this operation must be performed if the magnetic field of the drone or the earth's magnetic field (magnetic declination) changes.
No .. the calibration process identifies and measures the magnetic fields that are part of the drone as opposed to those that are outside of the drone (no matter what or where they are)

I can't do anything anymore. I give up.

It's math, if you change a variable in an equation the result is no longer true and needs to be corrected.

LV_Forestry Posted at 4-13 05:31
I can't do anything anymore. I give up.

It's math, if you change a variable in an equation the result is no longer true and needs to be corrected.

What do you mean?
Are you still on about mag declination?
I've given you all the information you should need to understand.

Your drone doesn't have to fly from Riga to Oslo.
Magnetic declination isn't relevant to it and your compass doesn't need recalibration.

Labroides Posted at 4-13 05:37
What do you mean?
Are you still on about mag declination?
I've given you all the information you should need to understand.

Lab, you won't make me change my mind that easily.

I could be wrong, it's far from impossible. But if only DJI was the only one to offer this calibration. PixHawk & Ardupilot (+/- the same), but above all the manufacturer of the SBG inertial units of my most advanced sensors, far from the DJI range. The guys are the masters of the magnetometer. I can't see myself going to ask them why I have to do the calibration before each flight, simply because I read a thread on the Mavic forum. Interesting thread but contradicts itself in its conclusion.

Besides, about the SBG Units before, it was necessary to perform a pattern to calibrate them, on the ground or in flight. Now it's possible to do it on the ground, but it's done automatically in flight.

I am attaching some screenshots so that you realize that this procedure is not trivial and highly recommended in high end equipment.

LV_Forestry Posted at 4-13 06:09
Lab, you won't make me change my mind that easily.

I could be wrong, it's far from impossible. But if only DJI was the only one to offer this calibration. PixHawk & Ardupilot (+/- the same), but above all the manufacturer of the SBG inertial units of my most advanced sensors, far from the DJI range. The guys are the masters of the magnetometer. I can't see myself going to ask them why I have to do the calibration before each flight, simply because I read a thread on the Mavic forum. Interesting thread but contradicts itself in its conclusion.

Lab, you won't make me change my mind that easily.
I'm not going to get into a back and forth with someone who can't or won't understand facts when they are presented on a plate.

I could be wrong
You are

But if only DJI was the only one to offer this calibration. PixHawk & Ardupilot (+/- the same), but above all the manufacturer of the SBG inertial units of my most advanced sensors, far from the DJI range. The guys are the masters of the magnetometer. I can't see myself going to ask them why I have to do the calibration before each flight, simply because I read a thread on the Mavic forum.
Sorry ... I cannot understand that mumbo jumbo.

I'll close with this ..
I've used one drone extensively in hard professional use for over 5 years without calibrating anything ,ever.
In that time I've travelled 5000 km east-west and >3000 km north-south.
The drone flew as well as it did straight out of the box.
My current working drone has travelled overseas too and flys perfectly too without callibration.
DJI's manual finally has factual advice and advises to only recalibrate if the drone needs it.

Interesting thread but contradicts itself in its conclusion.
It doesn't, your reading must be faulty.

The reference I showed you is an authoritative piece written by a very highly qualified scientist and expert drone flyer.
I cannot understand how you can read it and not accept it.
That's on you, not the material.

I'm sick of back and forth nonsense with people that prefer myths to facts.
Don't bother me any more on this topic, I've given you all I can.
If you don't want to believe it, I don't care.
I'm sorry I wasted my time on you.

Over & Out

Labroides Posted at 4-13 06:26
Lab, you won't make me change my mind that easily.
I'm not going to get into a back and forth with someone who can't or won't understand facts when they are presented on a plate.

I'm sorry you take it like that.

But if you don't want to waste time then don't come to a forum. Create a blog "I Labroides know everything and explain it to you and your opinion doesn't matter to me" and then voila!

LV_Forestry Posted at 4-13 06:53
I'm sorry you take it like that.

But if you don't want to waste time then don't come to a forum. Create a blog "I Labroides know everything and explain it to you and your opinion doesn't matter to me" and then voila!

Get lost ... I had enough of this crap yesterday.
Unfortunately there are very few flyers in this forum with any technical knowledge.

But if you don't want to waste time then don't come to a forum.
I come here to help flyers, but it's very hard when they are like you (and a few others here).
I've been helping members sort out flight problems and find lost drones since 2014, but lately it's getting harder.

You read what I gave you and still believe that nonsense about magnetic declination.
What's wrong with you?

Labroides Posted at 4-13 07:10
Get lost ... I had enough of this crap yesterday.
Unfortunately there are very few flyers in this forum with any technical knowledge.
You read what I gave you and still believe that nonsense about magnetic declination.

In one hand I have my experience with the integration of magnetometer in precise measurement systems, and in the other I have a text written without source by a stranger on the internet and recommended by another stranger still on the internet.

What is the most rational path for me to take?

Also, from my side I find the discussion interesting. I'm not saying I'm sure I'm right. I took note of what you shared. Be sure that at the next seminar at the university or elsewhere, if I meet specialists I will not fail to ask them the question.

Labroides Posted at 4-13 03:24
That was beautiful...
But it's completely irrelevant to flying your DJI drone.

Well off course it is completely irrelevant to fly your DJI drone...

He was talking about you... What's wrong? Don't you recognise sarcasm anymore??

That's why it's beautiful...

Hi there. Thank you for reaching out. Calibration is important in DJI drones to ensure that the drone's sensors and control systems are functioning correctly and accurately. Calibration is the process of setting the drone's sensors and controls to a known state, allowing them to accurately interpret and respond to input.

There are several sensors and systems that require calibration in DJI drones, including the compass, IMU (inertial measurement unit), and gimbal. The compass is responsible for providing the drone with its heading and direction, while the IMU measures the drone's acceleration, orientation, and rotation. The gimbal is responsible for stabilizing the camera and ensuring that it remains level and steady during flight. It is important to calibrate the drone regularly to ensure the stability of the unit. Should you have other inquiries, please don't hesitate to reach us. Thank you for your valued support.

Thank You everyone for your input but this seems to have got way out of hand.   There was no need for anyone to get aggressive over a question that only needed a simple answer and I can only apologise for raising such a contentious subject.......

Ideally I was after an official answer from DJI and to an extent Paladin has provided that for me but I have taken on board all other members input so please don't think I am dismissing any of your information or comments as there was a wealth of information supplied which I found extremely interesting.

Thank You, stay cheerful folks and enjoy your flying

djiuser_eFdo5aGvsfJ3 Posted at 4-14 02:04
Thank You everyone for your input but this seems to have got way out of hand.   There was no need for anyone to get aggressive over a question that only needed a simple answer and I can only apologise for raising such a contentious subject.......

Ideally I was after an official answer from DJI and to an extent Paladin has provided that for me but I have taken on board all other members input so please don't think I am dismissing any of your information or comments as there was a wealth of information supplied which I found extremely interesting.

Thank You everyone for your input but this seems to have got way out of hand.
It only got out of hand because one member hijacked the thread and took over posting misinformation.
Don't be mislead by his ignorance.

There was no need for anyone to get aggressive over a question that only needed a simple answer and I can only apologise for raising such a contentious subject...
It's not a contentious subject at all ... unless you are one of the few out there hang on to old and disproven myths.

Ideally I was after an official answer from DJI and to an extent Paladin has provided that for me.
Unfortunately the moderators here don't fly and have zero technical knowledge.
They just copy and past without any understanding.
What he posted is just vague generic information.

The link I provided is the most authoratitive information available on the topic.

I too have "authoritative information"


Today I made screenshots of what the application does to me every day. I tried with Phantom 4 and Mavic 2.


I went this morning to a site to do a survey, calibration of the compass as usual before anything else.


Then in the afternoon I went to another site, 64km away. And the surprise:



Phantom 4:


Mavic 2:


Right in the middle of a forest there are no houses, no metal structures, no nearby GSM relays, no wifi relays... So ok, maybe it's not the magnetic declination.
But then what is it? The Holy Spirit ?

I feel like this has absolutely nothing to do with flying a DJI drone…

I tried to calibrate the imu but it always failed

Are you orientating  the drone precisely as it is depicted at each stage of the caibration process? The required orientation changes at each stage.

djiuser_mMw8a934R4ga Posted at 9-30 07:01
I tried to calibrate the imu but it always failed

Are you orientating  the drone precisely as it is depicted at each stage of the caibration process? The required orientation changes at each stage.