Sudden compass error caused crash at take off

[Mycamblur]

After 7 years of flying, I have broken my first blade.

Immediately after take off, the drone (Mavic Pro) switched itself into 'atti mode' and within seconds had hurled itself backwards into a wall. Once on the floor, I saw that it was once more in GPS mode. No flight possible though, as blade broken with no spare.... and in a stunning location in perfect weather (zero wind).

I will change the blade when home, hopefully there is no other damage. None visible at least.

So...

1. What causes 'compass errors', and

2. is this what caused it to go into atti mode?

3. Why did atti mode make the drone fly backwards (at speed) at a pitched angle out of control, considering there was no wind?

Flight record attached and link

Could have been worse I guess and ended in water if it had gone in the other direction.

DJI Flight Log Viewer - PhantomHelp.com

Thanks guys

 

[Meta4]

3. Why did atti mode make the drone fly backwards (at speed) at a pitched angle out of control, considering there was no wind?

The speed was quite slow, maxing out at <2 metres/sec.
The pitch angle was a gentle 15°.
Atti Mode doesn't cause the drone to fly off uncontrolled.
In this incident, the drone being in atti mode is a by-product of the compass error.

Sudden uncommanded flight is most often a result of a yaw error , caused by powering up somewhere that magnetic interference deflects the compass.
The compass is where the gyro sensor in the IMU gets its initial directional data from.
The compass returns to normal as the drone moves away from the magnetic interference, but the gyro continues to have incorrect directional data.

I can't see the characteristic signs of a yaw error in your data so I'm not certain of the cause.

 

[trevsdad]

After 7 years of flying, I have broken my first blade.

Immediately after take off, the drone (Mavic Pro) switched itself into 'atti mode' and within seconds had hurled itself backwards into a wall. Once on the floor, I saw that it was once more in GPS mode. No flight possible though, as blade broken with no spare.... and in a stunning location in perfect weather (zero wind).

I will change the blade when home, hopefully there is no other damage. None visible at least.

So...

1. What causes 'compass errors', and

2. is this what caused it to go into atti mode?

3. Why did atti mode make the drone fly backwards (at speed) at a pitched angle out of control, considering there was no wind?

Flight record attached and link

Could have been worse I guess and ended in water if it had gone in the other direction.

DJI Flight Log Viewer - PhantomHelp.com

Thanks guys

What kind of surface did you launch from?

 

[New England Droning]

I'm watching this because after the big update earlier this week, the next flight I was looking to do a sunset launch. I've been doing these for years from my wooden deck. I will usually get all my sats and then start/takeoff but on many occasions I take off with 4-9 sats, climb to 3-4 ft and move away from the building to get the rest of the sats.

This mission I had 5 sats and took off and at 6-8 inches it suddenly went backward into a deck chair and screen... After one year I had my first crash with the mavic 3 Pro. A couple of scratched blades checked and flew a mission over my house. I enjoy my sunsets...
So I'll see if happens to others...

 

[MS Coast]

I'm watching this because after the big update earlier this week, the next flight I was looking to do a sunset launch. I've been doing these for years from my wooden deck. I will usually get all my sats and then start/takeoff but on many occasions I take off with 4-9 sats, climb to 3-4 ft and move away from the building to get the rest of the sats.

This mission I had 5 sats and took off and at 6-8 inches it suddenly went backward into a deck chair and screen... After one year I had my first crash with the mavic 3 Pro. A couple of scratched blades checked and flew a mission over my house. I enjoy my sunsets...
So I'll see if happens to others...

Risky practices.

Did you check to see that the drone's compass was properly set before launching? It may well have been biased by nails, screws, or attachment hardware in your deck and caused that behavior.

Flight logs?

 

[EricJT]

Risky practices.

Did you check to see that the drone's compass was properly set before launching? It may well have been biased by nails, screws, or attachment hardware in your deck and caused that behavior.

Flight logs?

When I take off from the dock at the cottage, I hold my M2P at shoulder level until it does its power up routine and GPS acquisition. Making sure the compass on the RC is pointing in the same direction as the drone, l place it on the dock and then take off.

I don't trust taking off from my hand....I need all my fingers intact.

 

[Meta4]

This mission I had 5 sats and took off and at 6-8 inches it suddenly went backward into a deck chair and screen... After one year I had my first crash with the mavic 3 Pro. A couple of scratched blades checked and flew a mission over my house. I enjoy my sunsets...

This sounds very much like a yaw error and it should be easy to confirm if you post your flight data.
There are a couple of options ...

1. Go to DJI Flight Log Viewer | Phantom Help
Follow the instructions there to upload your flight record from your phone or tablet.
That will give you a detailed report on the flight data.
Come back and post a link to the report it provides and someone might be able to analyse it and give you an understanding of the cause of the incident.
or
2. Just post the .txt file here
or
3. If you use Airdata, you can view the flight data on Airdata and post a link for the Airdata report


If you are using one of the controllers with an integrated screen, the txt files are to be found here:

Android\data\dji.go.v5\files\FlightRecord

So I'll see if happens to others...

Your issue had nothing at all to do with the update.
If it did, hundreds of other flyers would have been similarly affected.

 

[MS Coast]

When I take off from the dock at the cottage, I hold my M2P at shoulder level until it does its power up routine and GPS acquisition. Making sure the compass on the RC is pointing in the same direction as the drone, l place it on the dock and then take off.

I don't trust taking off from my hand....I need all my fingers intact.

Check them against the reality, rather than one another. The controller gets its initial orientation from the compass, so if the compass is biased, the controller will be, too.

 

[Wild Drone Pilot]

After 7 years of flying, I have broken my first blade.

Immediately after take off, the drone (Mavic Pro) switched itself into 'atti mode' and within seconds had hurled itself backwards into a wall. Once on the floor, I saw that it was once more in GPS mode. No flight possible though, as blade broken with no spare.... and in a stunning location in perfect weather (zero wind).

I will change the blade when home, hopefully there is no other damage. None visible at least.

So...

1. What causes 'compass errors', and

2. is this what caused it to go into atti mode?

3. Why did atti mode make the drone fly backwards (at speed) at a pitched angle out of control, considering there was no wind?

Flight record attached and link

Could have been worse I guess and ended in water if it had gone in the other direction.

DJI Flight Log Viewer - PhantomHelp.com

Thanks guys

Around any metal?
Did it for me has a name I think it is washbowl effect? My 1st time was my DJI Spark Almost Crashed

 

[EricJT]

Check them against the reality, rather than one another. The controller gets its initial orientation from the compass, so if the compass is biased, the controller will be, too.

Thanks for your comment....I failed to include I know I am facing south and the RC and drone must agree with that direction.

 

[jwilson]

I've never done a compass calibration in my life. I don't know why a drone needs a compass. GPS should be enough.

 

[Meta4]

I've never done a compass calibration in my life.

That's completely normal.
You'd only need to recalibrate the compass if you add or remove accessories to/from the drone.
(That's accessories that could have an effect on the magnetic fields of the drone.)

I don't know why a drone needs a compass. GPS should be enough.

A GPS has no way to sense direction.
Without sensors to provide directional information, the drone has no idea which way it is pointing or how to get to home or any other point.

 

[New England Droning]

This sounds very much like a yaw error and it should be easy to confirm if you post your flight data.
There are a couple of options ...

1. Go to DJI Flight Log Viewer | Phantom Help
Follow the instructions there to upload your flight record from your phone or tablet.
That will give you a detailed report on the flight data.
Come back and post a link to the report it provides and someone might be able to analyse it and give you an understanding of the cause of the incident.
or
2. Just post the .txt file here
or
3. If you use Airdata, you can view the flight data on Airdata and post a link for the Airdata report


If you are using one of the controllers with an integrated screen, the txt files are to be found here:

Android\data\dji.go.v5\files\FlightRecord



Your issue had nothing at all to do with the update.
If it did, hundreds of other flyers would have been similarly affected.

DJI Flight Log Viewer - PhantomHelp.com

I believe this is the file you are looking for.
I see nothing but user error. I "should" have waited for a full complement of sats. I pulled back ( on the right-hand controller. But in my defence this was in response to it lurching backward. I went into ATTI mode twice but this is to be expected since it didn't have a full complement of Sats (7-10 sats) depending on when you look.
but at 8.8 sec I was was pulling slightly on the right hand stick which probably caused the backward motion and then I hit it hard and that sealed my fate.
So basically I have heavy fingers and need to NOT touch the RH controller...

Am I interpreting what I'm seeing in the log correctly?

 

[Meta4]

Am I interpreting what I'm seeing in the log correctly?

That's about right.
The light level was too low for VPS to help hold position and your GPS reliability was at 0/5 for the whole short flight.
Launching before getting good GPS location data, isn't a problem .. if there are no obstacles to collide with.
But with obstacles close by and no obstacle avoidance, full stick moves are to be avoided.

 

[jwilson]

That's completely normal.
You'd only need to recalibrate the compass if you add or remove accessories to/from the drone.
(That's accessories that could have an effect on the magnetic fields of the drone.)


A GPS has no way to sense direction.
Without sensors to provide directional information, the drone has no idea which way it is pointing or how to get to home or any other point.

So are you saying that if drones didn't have a compass, they wouldn't work? I think if you push return to home, it's going to get to your home point, with or without a compass. Because as soon as it moves, it can tell if the home point is closer or further, and can yaw accordingly. If you have two points in space, you can calculate your direction on a 2 dimensional horizontal plane.

 

[Meta4]

So are you saying that if drones didn't have a compass, they wouldn't work? I think if you push return to home, it's going to get to your home point, with or without a compass. Because as soon as it moves, it can tell if the home point is closer or further, and can yaw accordingly. If you have two points in space, you can calculate your direction on a 2 dimensional horizontal plane.

Perhaps you should tell DJI that their drones don't need compasses then.

 

[New England Droning]

That's about right.
The light level was too low for VPS to help hold position and your GPS reliability was at 0/5 for the whole short flight.
Launching before getting good GPS location data, isn't a problem .. if there are no obstacles to collide with.
But with obstacles close by and no obstacle avoidance, full stick moves are to be avoided.

As always, I'm my own worst enemy...

 

[EricJT]

Perhaps you should tell DJI that their drones don't need compasses then.

Meta4, I tried looking for your past excellent post regarding how the FC, IMU & compass get initiated and interact properly upon startup after the required number of GPS satellites are obtained....I couldn't find it.

I think if you reposted it here it might educate more of the forum readers.

 

[Meta4]

Meta4, I tried looking for your past excellent post regarding how the FC, IMU & compass get initiated and interact properly upon startup after the required number of GPS satellites are obtained....I couldn't find it.

I think if you reposted it here it might educate more of the forum readers.

The IMU gets its initial direction data from the compass as soon as the drone is powered on.
The compass etc is independent of GPS.

The post you are looking for isn't mine (but I quote it a lot).

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

 

[RodPad]

So are you saying that if drones didn't have a compass, they wouldn't work? I think if you push return to home, it's going to get to your home point, with or without a compass. Because as soon as it moves, it can tell if the home point is closer or further, and can yaw accordingly. If you have two points in space, you can calculate your direction on a 2 dimensional horizontal plane.

Complected thread to follow.
Compass issues are a major factors in RTH.

Your statement of GPS should work with out a compass RTH I agree!.
Reading many post, and viewing lots of flight logs.

Not on consumer models at this time, other levels I would assume yes.

Rod ..