First Flight in the U.S Crashed my M3

[redserv]

Hello Guys
Unfortunately the day has come where i have my first major crash, and its my Mavic 3. I am now living in the U.S and decided to fly for the first time from my back yard in Florida.
I always hand launch my bird acquiring 10+ Sats first. For some reason this time i launched from the ground and noticed sats taking longer than normal to aquire HP. At 8 sats I decided to launch my bird and fly up to get the rest of sats. I looked at my drone and she took a hard flight to the right and crashed in some bushes at the side of the road. At this point i just want to confirm that it was not anything else. I will appreciate the experts here taking a look at my attached logs. Thanks in advance guys

 

[mavic3usa]

"At 8 sats I decided to launch my bird and fly up to get the rest of sats."

I haven't heard of this, is this some sort of technique that needs to be applied here in America that you didn't need back at home in your previous country?

 

[redserv]

"At 8 sats I decided to launch my bird and fly up to get the rest of sats."

I haven't heard of this, is this some sort of technique that needs to be applied here in America that you didn't need back at home in your previous country?

No it was an error in judgement. No need to get smart.

 

[mavic3usa]

No it was an error in judgement. No need to get smart.

Sorry, no offense meant. I read too much into the whole first time in America part of the complaint. I'm sure if you go ahead and attach your logs, someone here could take a look for you. Hope your drone is ok and welcome to America!

 

[Meta4]

I always hand launch my bird acquiring 10+ Sats first. For some reason this time i launched from the ground and noticed sats taking longer than normal to aquire HP.

That's normal when you are in a new location or you haven't flown for a few weeks.
This explains it:

Time To First Fix – Hot Start, Warm Start, Cold Start. -

Time to First Fix (commonly abbreviated to TTFF) is the process of time in which it takes a GPS enabled device to obtain signal from enough satellites to determine accurately its location. A GPS receiver must have three sets of data before it can accurately locate itself. GPS satellite signals...

gpsbeam.com

Your drone will take longer still if buildings, trees etc are blocking the skyview.

At 8 sats I decided to launch my bird and fly up to get the rest of sats.

As the Mavic 3 is receiving sats from 3 separate gps systems, having 8 sats is almost none.
10 is not much better.
Instead of thinking there's a magic number, you need to wait until the sat icon on the screen turns white and drone records its home point.

I will appreciate the experts here taking a look at my attached logs. Thanks in advance guys

It would help if you posted the logs.

 

[redserv]

That's normal when you are in a new location or you haven't flown for a few weeks.
This explains it:

Time To First Fix – Hot Start, Warm Start, Cold Start. -

Time to First Fix (commonly abbreviated to TTFF) is the process of time in which it takes a GPS enabled device to obtain signal from enough satellites to determine accurately its location. A GPS receiver must have three sets of data before it can accurately locate itself. GPS satellite signals...

gpsbeam.com

Your drone will take longer still if buildings, trees etc are blocking the skyview.

As the Mavic 3 is receiving sats from 3 separate gps systems, having 8 sats is almost none.
10 is not much better.
Instead of thinking there's a magic number, you need to wait until the sat icon on the screen turns white and drone records its home point.

It would help if you posted the logs.

Thank you Meta for your advice. HonestlyI think I became impatient and paid the price. Although when i launched the drone at 8 sats i went up in altitude an think i heard HP updated. Anyway, I will upload the logs so you can take a look at it, Thanks again for your assistance.

 

[redserv]

Here is the last log

 

[Meta4]

Here is the last log

Your incident is unusual and has some of the indications of a yaw error, but not enough to tell that this was the cause.
AT 3:38.3 the drone started to pick up speed without any matching joystick input.
Within a few seconds, the speed was quite high.

Do you remember what direction the drone was pointing when it was on the ground at launch time?

 

[redserv]

Your incident is unusual and has some of the indications of a yaw error, but not enough to tell that this was the cause.
AT 3:38.3 the drone started to pick up speed without any matching joystick input.
Within a few seconds, the speed was quite high.

Do you remember what direction the drone was pointing when it was on the ground at launch time?

Hello Meta thanks for looking at the logs. The drone was facing Southwest and crashed going sideways heading northwest in some bushes

 

[Meta4]

Hello Meta thanks for looking at the logs. The drone was facing Southwest and crashed going sideways heading northwest in some bushes

OK ... it was a yaw error that caused the issue.
The data indicates the drone was pointing just a little north of east until the trouble started.
It will have been caused by powering up in an area of magnetic interference.
On or very close to reinforced concrete or something made of steel perhaps.
That deflected the compass which was used to give the gyro sensors their initial directional information.
The compass would have returned to correct readings after it left the magnetic field, but the gyro would have been reading incorrectly
While the drone didn't have good GPS data, it wasn't a problem.
But once GPS was available, the data from the compass and gyro would have been in disagreement.
The drone sensed movement and tried to correct, but using incorrect heading data from the gyro sensor, it moved the wrong way.
It would have kept doing that many times per second and getting further away from where it was trying to go each time.
That's why it accelerated so fast.

 

[redserv]

OK ... it was a yaw error that caused the issue.
The data indicates the drone was pointing just a little north of east until the trouble started.
It will have been caused by powering up in an area of magnetic interference.
On or very close to reinforced concrete or something made of steel perhaps.
That deflected the compass which was used to give the gyro sensors their initial directional information.
The compass would have returned to correct readings after it left the magnetic field, but the gyro would have been reading incorrectly
While the drone didn't have good GPS data, it wasn't a problem.
But once GPS was available, the data from the compass and gyro would have been in disagreement.
The drone sensed movement and tried to correct, but using incorrect heading data from the gyro sensor, it moved the wrong way.
It would have kept doing that many times per second and getting further away from where it was trying to go each time.
That's why it accelerated so fast.

Thank you so much Meta. Yup I launched from a a tiled area that probably was reinforced with steel. smh.... Its funny one of my pre flight checks is to make sure I dont have any compass warning etc.. This is the first time since I have owned the Mavic 3 that i did NOT hand launch. Dunno if it would have helped or not. Oh well, lessons learned. Hopefully i can get my drone repaired soon. Thank you so much Meta, much appreciated.

 

[Meta4]

Its funny one of my pre flight checks is to make sure I dont have any compass warning etc..

Your compass would only warn if the magnetic field is significantly stronger than the earth's normal magnetic field.
But if the problem field is a similar strength (but a different direction), that won't set off alarm bells.
The way to check that this isn't happening is to look at the drone icon on the map display and ensure that the icon is pointing the same direction as the actual drone is.
In this incident the drone was pointing SW but the icon on the map display would have been pointing 75° or just north of east.

 

[rwilliam99]

This is why I try and read all of these posts - to learn things so I can not crash my drone again. The practice of ensuring that the drone is pointing in the correct direction on the maps was something new for me - I had heard it but it seemed odd to me until I understood WHY it was important - I now do that on all my flights to make sure there are no compass issues.

 

[redserv]

Your compass would only warn if the magnetic field is significantly stronger than the earth's normal magnetic field.
But if the problem field is a similar strength (but a different direction), that won't set off alarm bells.
The way to check that this isn't happening is to look at the drone icon on the map display and ensure that the icon is pointing the same direction as the actual drone is.
In this incident the drone was pointing SW but the icon on the map display would have been pointing 75° or just north of east.

Very interesting. Then I definitely missed it. What was evident as well is the drone was not showing the map in the left corner before take off. I am going to be checking that the drone is pointing in the same direction from now on. I have 58 flights with the M3 and this is the first time I haven't launched. I do this to Hopefully prevent the compass issue that I got today. Thanks again for your effort and explaining where I went wrong. Thank god my drone was repaired today by me.

 

[KB9Radio]

Very interesting. Then I definitely missed it. What was evident as well is the drone was not showing the map in the left corner before take off. I am going to be checking that the drone is pointing in the same direction from now on. I have 58 flights with the M3 and this is the first time I haven't launched. I do this to Hopefully prevent the compass issue that I got today. Thanks again for your effort and explaining where I went wrong. Thank god my drone was repaired today by me.

Pickup a landing pad sometime, then you can launch from the grass/dirt or even snow, which is usually safe. Hard surfaces such as concrete, tile on concrete, are normally reinforced with wire or rebar, also vehicle surfaces, all of which will mess with the compass.

 

[JoelP]

OK ... it was a yaw error that caused the issue.
The data indicates the drone was pointing just a little north of east until the trouble started.
It will have been caused by powering up in an area of magnetic interference.
On or very close to reinforced concrete or something made of steel perhaps.
That deflected the compass which was used to give the gyro sensors their initial directional information.
The compass would have returned to correct readings after it left the magnetic field, but the gyro would have been reading incorrectly
While the drone didn't have good GPS data, it wasn't a problem.
But once GPS was available, the data from the compass and gyro would have been in disagreement.
The drone sensed movement and tried to correct, but using incorrect heading data from the gyro sensor, it moved the wrong way.
It would have kept doing that many times per second and getting further away from where it was trying to go each time.
That's why it accelerated so fast.

It was very kind of you, Meta4, to analyze the flight data. Hopefully, others here will make it a habit to check the orientation on their map versus reality before taking off.

That said, what is the remedy when one inadvertently makes this error as an oversight and finds their drone in the air with a compass error. Is there a safe recovery?

 

[redserv]

I was really lucky that my drone survived. I got some minor scuff marks. I changed all my props and was able to reattach the vibration rubbers back on the gimbal. I was expecting the worse to be honest. I will be talking the drone for the first test flight today. I also did a IMU and will be doing a compass calibration at the park. I am thankful as the outcome could have been worse, slot worse!

 

[Felix le Chat]

Hello Guys
Unfortunately the day has come where i have my first major crash, and its my Mavic 3. I am now living in the U.S and decided to fly for the first time from my back yard in Florida.
I always hand launch my bird acquiring 10+ Sats first. For some reason this time i launched from the ground and noticed sats taking longer than normal to aquire HP. At 8 sats I decided to launch my bird and fly up to get the rest of sats. I looked at my drone and she took a hard flight to the right and crashed in some bushes at the side of the road. At this point i just want to confirm that it was not anything else. I will appreciate the experts here taking a look at my attached logs. Thanks in advance guys

You're not on your own. I've noticed something similar. Mini 3 pro: booted up on the deck prior to hand launch seems to take considerable time acquiring between 8 and 11 sats and then fails to register more, after I put her up in the air and hover at just over head height, the locked sats speedily increase to between 28 and 31..... THEN I'll send her for a flight

 

[redserv]

You're not on your own. I've noticed something similar. Mini 3 pro: booted up on the deck prior to hand launch seems to take considerable time acquiring between 8 and 11 sats and then fails to register more, after I put her up in the air and hover at just over head height, the locked sats speedily increase to between 28 and 31..... THEN I'll send her for a flight

I was standing under a tree and flying in a new location was the reason for taking longer to acquire GPS. I should have known this as I took off with only. 8 sats. Although that was not issue that lead the crash entirely. I think I need to do a better job at pre flight tests before take off. The things Meta said for example checking the orientation of the drone in relation to the map was not the first time I heard this from him. I just did not put it into practice. My accident could have been worse and avoided. Also this one of the reason I hand launch almost all the time, to prevent the dreaded yaw error. Lessons for us all

 

[Meta4]

That said, what is the remedy when one inadvertently makes this error as an oversight and finds their drone in the air with a compass error. Is there a safe recovery?

You could be lucky and only have a small yaw error where the difference between the gyro and actual is only a small angle.
In that case you could get the drone back.
Unfortunately these are uncommon and the outcome is more often, a drone launching itself sideways on a curved path at crazy speed and completely uncontrollable.
That sort usually end quickly with a crash.