NEO Crash Help

[Chrislaf]

Today went out to fly my NEO. Was using my RC-N3 controller with my iPhone. I launched from my hand out on the ice of a frozen lake just out from the beach. As soon as the NEO lifted off it took off backward at high speed and likely crashed into a pole or a cement barrier. It broke a propellor but otherwise was fine ( a few tiny scuff marks).

I would like to figure out what happened so it doesn't happen again. I suspect it was a compass error caused by the rechargeable hand warmers that were in my coat pocket. They have a magnet that attaches the two of them together to create heat on both sides. The other possibility is that there is something wrong with the sticks on my RC-N3 controller which is unlikely. I should have checked the AC orientation as I usually do but today I didn't.

Here is the log file NEO Crash log

Thanks for any help.

Chris

 

[msinger]

I launched from my hand out on the ice of a frozen lake just out from the beach

Where was the Neo when it was powered on?

I should have checked the AC orientation as I usually do but today I didn't

Is this the direction the drone was pointed at takeoff?

 

[Chrislaf]

Thanks for your reply.

Where was the Neo when it was powered on?

It was powered on in my hand.

Is this the direction the drone was pointed at takeoff?

View attachment 180795

I'm pretty sure I had the NEO facing me opposite the direction shown in the log.

The rechargeable batteries, two of them attach together via a magnet. I had them in my left pocket and was holding the NEO in my right hand with the controller cradled over my left arm. I powered on the NEO with my right hand.

Chris

 

[Chrislaf]

I was out flying the NEO yesterday in same location with the rechargeable hand warmers. The only difference was I launched from a landing pad that was on the ice.

Chris

 

[slup]

...I suspect it was a compass error...

Yeah... the available data strongly indicate just that.

The IMU got initialized about 180 degrees wrong, which quickly caused a straight fly away as soon as the IMU tried to hold position. The drone quickly reached max tilt & speed as the flight controller applied thrust with the wrong motors due to that the positional error just became larger instead of coming back to the commanded position. Looking at the deviation between the IMU & GPS velocity tells the same story.

In some way you powered on the drone in magnetic disturbance... which deflected the compass 180 degrees, this could have been avoided if you had followed your ordinary drill & checked the drone direction on the map.

 

[samysys]

Today went out to fly my NEO. Was using my RC-N3 controller with my iPhone. I launched from my hand out on the ice of a frozen lake just out from the beach. As soon as the NEO lifted off it took off backward at high speed and likely crashed into a pole or a cement barrier. It broke a propellor but otherwise was fine ( a few tiny scuff marks).

I would like to figure out what happened so it doesn't happen again. I suspect it was a compass error caused by the rechargeable hand warmers that were in my coat pocket. They have a magnet that attaches the two of them together to create heat on both sides. The other possibility is that there is something wrong with the sticks on my RC-N3 controller which is unlikely. I should have checked the AC orientation as I usually do but today I didn't.

Here is the log file NEO Crash log

Thanks for any help.

Chris

Hi,

I have a Neo too and I am not always comfortable with its flight control.

Anyway, I tried to reconstruct your case with some magnet. I conducted the test in a well lighted indoor condition. I put a rather strong magnet in different position of the Neo (front, across, back and top) before take off and none affect it's hovering. I even stick the magnet on top and under the Neo and it still fly alright. To my surprise, no warning for magnetic interference pop up.

I further restart the Neo with the magnet sticked to it and still have no effect on it's hovering....

I cannot rule out magnetic interference/ compass error in your case but I would say the chance is slim.

 

[slup]

...I conducted the test in a well lighted indoor condition.

I cannot rule out magnetic interference/ compass error in your case but I would say the chance is slim.

You need a good GPS lock & something that tries to move you drone out of position uncommanded (like wind) to experience the full effect of a yaw error... flying indoors "in a well lighted indoor condition" indicate that you didn't fully replicate the OP's conditions, your drone likely had a good VPS lock on the surface below it.

Additionally to the above, you need to deflect the compass just in the moment when the drone is powered on... moving the drone into magnetic disturbance after that will not create a yaw error fly away.

 

[Meta4]

I cannot rule out magnetic interference/ compass error in your case but I would say the chance is slim.

The data clearly shows that the incident was a yaw error caused by powering up where magnetic interference deflected the compass, initialising the gyro sensor with incorrect directional data.

 

[samysys]

You need a good GPS lock & something that tries to move you drone out of position uncommanded (like wind) to experience the full effect of a yaw error... flying indoors "in a well lighted indoor condition" indicate that you didn't fully replicate the OP's conditions, your drone likely had a good VPS lock on the surface below it.

Additionally to the above, you need to deflect the compass just in the moment when the drone is powered on... moving the drone into magnetic disturbance after that will not create a yaw error fly away.

thanks for your comment. I did it indoor as this was a more controlled environment. I thought DJI APP would pop up warning before the magnet had any actual effect on the drone. I did try to start the Neo with a magnet sticked to it but still no warning pop up.

 

[slup]

...I did it indoor as this was a more controlled environment. I thought DJI APP would pop up warning...

You can't draw any kind of conclusion by creating tests not fully understanding what's needed to cause a yaw error, & also change the circumstances substantially. Your belief about the app warning as a good indicator isn't correct either, it's not a good indicator at all when it comes to indicate a risk for a yaw error... it only let you know that the magnetic disturbance have reached over a set threshold, which usually is much higher that what's needed to deflect the compass enough to get a yaw error.

So going with that no warning have popped up, it shouldn't be a risk for a yaw error... is wrong. The only way is to compare the drone direction on the map with how the drone points in reality, if the map & reality directions deviate... power down, move to another spot, power up & compare again.

 

[Chrislaf]

Yeah... the available data strongly indicate just that.

The IMU got initialized about 180 degrees wrong, which quickly caused a straight fly away as soon as the IMU tried to hold position. The drone quickly reached max tilt & speed as the flight controller applied thrust with the wrong motors due to that the positional error just became larger instead of coming back to the commanded position. Looking at the deviation between the IMU & GPS velocity tells the same story.

In some way you powered on the drone in magnetic disturbance... which deflected the compass 180 degrees, this could have been avoided if you had followed your ordinary drill & checked the drone direction on the map.

Yeah, I usually always check the orientation of the drone on the map but it was cold and my fingers were getting cold and just wanted to get the NEO up so I could warm up my hands. Since I had been using the rechargeable hand warmers before, I became too complacent and launched without checking. Lesson learned.

Thanks for your reply.

Chris

 

[Chrislaf]

Thanks to all of you for replying and to @msinger, @slup and @Meta4 for confirming the yaw error. The only other possibility I was concerned about was some sort of problem with the controller sticks.

I had been concerned that the magnets in the hand warmers might be a problem and now I know, they are if I am not careful.

Chris

 

[Chrislaf]

I guess what surprised me about the hand warmers is how strong the magnet is. Is doesn't appear to be that strong since the two warmers slide around when held together by the magnet.

I will have to be especially careful when using the NEO with my phone as a controller or just with the control panel on the drone itself (rarely use), since there is no way to check the orientation of the drone because there is no map or radar display. Also, disposable hand warmers will also likely be a problem since they use iron powder to create heat if they are close to the drone when powered on.

Chris

 

[msinger]

The only other possibility I was concerned about was some sort of problem with the controller sticks.

I don't think I've ever seen an issue like that. In your case, it's easy to rule out since the drone is moving in parts of the flight without any stick input.

 

[Chrislaf]

I don't think I've ever seen an issue like that. In your case, it's easy to rule out since the drone is moving in parts of the flight without any stick input.

Yeah, I was looking at that and that's what I thought but I don't have enough knowledge about these things to interpret the log file properly. I am grateful to you and others for confirming my suspicions and ruling out a controller issue.

I will likely be resorting to using my landing pad or backpack whenever possible and checking the orientation on the map before I launch my NEO when the weather is cold like this since I need my hand warmers to keep my hands from freezing.

Chris

 

[msinger]

Yeah, I was looking at that and that's what I thought but I don't have enough knowledge about these things to interpret the log file properly

It might be easier to interpret the data if you graph it with a tool like Flight Reader. For example, you can see that the right stick stays mostly centered (at 1024), while the speed increases at a relatively consistent rate.




And then here you can see that the drone is yawing pretty consistently, even though the left stick isn't commanding it to do so:

 

[Chrislaf]

I will have to check that out in Flight Reader which I have and use to log all my flights. Still learning how to use all of its features.

Chris

 

[msinger]

You can create a chart from here:

 

[Chrislaf]

Thanks. I was able to get the charts you displayed above. I just have to get more used to the lingo. I am going to make it a habit of looking at the charts of the data for flights from all of my drones so I can get better at understanding the data.

Chris

 

[samysys]

You can't draw any kind of conclusion by creating tests not fully understanding what's needed to cause a yaw error, & also change the circumstances substantially. Your belief about the app warning as a good indicator isn't correct either, it's not a good indicator at all when it comes to indicate a risk for a yaw error... it only let you know that the magnetic disturbance have reached over a set threshold, which usually is much higher that what's needed to deflect the compass enough to get a yaw error.

So going with that no warning have popped up, it shouldn't be a risk for a yaw error... is wrong. The only way is to compare the drone direction on the map with how the drone points in reality, if the map & reality directions deviate... power down, move to another spot, power up & compare again.

thanks and another lesson learnt~