Mavic 2 took off forward instead of upward, then crashed into sidewalk

[marcorasi1960]

maybe?

Transfer files between your computer & Android device - Android Help

You can use your Google Account or a USB cable to move photos, music, and other files between your computer and Android device. Important: Some of these steps work only on Android 9.0 and up.

support.google.com

I tried that too. Instructions tell me:

How to use it​

1. Download the app.
2. Open AndroidFileTransfer.dmg.
3. Drag Android File Transfer to Applications.
4. Use the USB cable that came with your Android device and connect it to your Mac.
5. Double click Android File Transfer.
6. Browse the files and folders on your Android device and copy files.

Everything goes smoothly until step 5. When I double click Android File Transfer, I get message: "No Android Device Found. Please connect your Android device with a USB cable to get started."

Thanks for the lead, though!

​

​

 

[Sjkieffer]

Wrong mode?

 

[sar104]

I tried that too. Instructions tell me:

How to use it​

1. Download the app.
2. Open AndroidFileTransfer.dmg.
3. Drag Android File Transfer to Applications.
4. Use the USB cable that came with your Android device and connect it to your Mac.
5. Double click Android File Transfer.
6. Browse the files and folders on your Android device and copy files.

Everything goes smoothly until step 5. When I double click Android File Transfer, I get message: "No Android Device Found. Please connect your Android device with a USB cable to get started."

Thanks for the lead, though!

​

​

Macs will not mount the Android file system directly, and so Android File Transfer or similar is the correct solution. That's what I use with the SC, and it works fine. Are you connecting via the lower USB-C connector on the SC to a USB-A input on the Mac or USB hub? The upper USB-A connector will not work for this purpose, and nor will USB-C directly to USB-C on the Mac.

That should give you an interface that looks like this:

 

[marcorasi1960]

Macs will not mount the Android file system directly, and so Android File Transfer or similar is the correct solution. That's what I use with the SC, and it works fine. Are you connecting via the lower USB-C connector on the SC to a USB-A input on the Mac or USB hub? The upper USB-A connector will not work for this purpose, and nor will USB-C directly to USB-C on the Mac.

That should give you an interface that looks like this:

View attachment 130882

Thank for your help. I am connecting via the lower USB-C connector on the SC, as you say I should. But I am also connecting directly to USB-C on the Mac, as you say I should not. My MacBook Pro is fairly recent and only has USB-C ports. So, let me get this straight: you recommend that I buy a cable with USB-C at one end (the one going into the SC), and USB-A at the other end; and then hook the USB-A end to my laptop using a USB-A to USB-C adaptor? I will certainly try that if you can confirm that you believe that that would work. Let me know. Thanks!

 

[sar104]

Thank for your help. I am connecting via the lower USB-C connector on the SC, as you say I should. But I am also connecting directly to USB-C on the Mac, as you say I should not. My MacBook Pro is fairly recent and only has USB-C ports. So, let me get this straight: you recommend that I buy a cable with USB-C at one end (the one going into the SC), and USB-A at the other end; and then hook the USB-A end to my laptop using a USB-A to USB-C adaptor? I will certainly try that if you can confirm that you believe that that would work. Let me know. Thanks!

Yes - I tested that again today. USB-C to USB-C does not work. My MacBook Pro also has only USB-C inputs - you need to use a USB hub that has USB-A inputs. I'm connecting the SC via the hub.

 

[marcorasi1960]

Yes - I tested that again today. USB-C to USB-C does not work. My MacBook Pro also has only USB-C inputs - you need to use a USB hub that has USB-A inputs. I'm connecting the SC via the hub.

Thank you soooo much! I would have never figured this one out on my own. I ordered the USB hub, so in a week or so I should be able to post the flight records and understand what happened. P.S. After replacing the broken rotors, I tried to see if the damaged drone would take off (in an area where it could do no damage). A message "Cannot take off" popped up, and it said something like: "The angle of inclination is too great, place the drone on a flat surface". It was on perfectly flat ground. I ordered a new Mavic 2

 

[Meta4]

A message "Cannot take off" popped up, and it said something like: "The angle of inclination is too great, place the drone on a flat surface". It was on perfectly flat ground.

You should recalibrate the IMU to get its sensors working properly again.

 

[marcorasi1960]

You should recalibrate the IMU to get its sensors working properly again.

Thanks, I will try that

 

[sar104]

Thank you soooo much! I would have never figured this one out on my own. I ordered the USB hub, so in a week or so I should be able to post the flight records and understand what happened. P.S. After replacing the broken rotors, I tried to see if the damaged drone would take off (in an area where it could do no damage). A message "Cannot take off" popped up, and it said something like: "The angle of inclination is too great, place the drone on a flat surface". It was on perfectly flat ground. I ordered a new Mavic 2

Okay - that should work. I'll edit the log retrieval section of the guide to make that clearer.

 

[Arnold LeVine]

Also check to see if you hear the cooling fan when you start the drone. The fan could have been what gave off the burning smell, and could cook your cpu. If you do not hear the fan, stop and get a repair done.

 

[trevsdad]

Without the Logs , There is a good chance that Magnetic Interference did take place, and hammer the Compass .
If you would have your Position on the Second flight you might have been ok,

Pavement is always sketchy , as is Concrete , best to take off in the Grass or hand Launch .
The only real mistake was the second flight in the exact same place.

Phantomrain.org
Gear to fly your Mavic 2 Zoom in the Rain.

I always reccomend to my students that they avoid taking off from a concrete surface due to potential problems from rebar. Don't know if the pavement the OP described was this type

 

[Arnold LeVine]

Opened my Smart Controller, no AC. clicked on the apps button (lower right corner of screen) opened file manager.
clicked on Internal Storage at upper right of screen. chose internal storage instead of Lexar SD card (which is what I have in the micro sd slot of the SC) opened DJI folder, scroll to dji.go.v4 folder and opened it, scroll to flight record folder and opened it. scroll past folders and I found a bunch of DJIFlightRecord_xxdatesetc..txt. Using a long-click-hold, "copy" as well as other functions appear at the bottom of screen. I clicked "copy" and it gave me the option of copy to the microSD card. Popped it out and was able to read via usb card reader. I assume this works for all the other files mentioned above, and for a Mac without any hubs or special software.

Will someone with more experience/knowledge than I please either verity what I am proposing here, or tell me why I am wrong and to ignore this workaround

 

[marcorasi1960]

Okay - that should work. I'll edit the log retrieval section of the guide to make that clearer.

Here are the two flight records of the 2 crashes I described. They are very small files because the flights lasted only a few seconds. Everything worked smoothly in finding these files once I got the USBC-USB plus the USB bank 'adaptor', just as you had told me. Are these in the required format, or should I load something different? If they are in the right format, any help in figuring out what happened will be greatly appreciated. I received my new Mavic-2 zoom, and as soon as I know how I destroyed the old one (i.e., what not to do) I will have the confidence to be safe in flight again. Thanks!!

 

[marcorasi1960]

Here are the two flight records of the 2 crashes I described. They are very small files because the flights lasted only a few seconds. Everything worked smoothly in finding these files once I got the USBC-USB plus the USB bank 'adaptor', just as you had told me. Are these in the required format, or should I load something different? If they are in the right format, any help in figuring out what happened will be greatly appreciated. I received my new Mavic-2 zoom, and as soon as I know how I destroyed the old one (i.e., what not to do) I will have the confidence to be safe in flight again. Thanks!!

I just re-read the Flight log retrieval and analysis guide, and realized that I should have probably loaded the DAT files. So here they are. I attached 5 of them, because I am not certain which two are the relevant ones. I believe they are the first 2, but I am not 100% sure, because the hour of the flight is clearly wrong, possibly due to the time zone difference between the US (where I usually fly) and Italy (where I am now). Thanks again!

 

[slup]

After looking into the logs it's easy to understand why the AC set off forward ... Something is clearly wrong with how both the IMU's think's the AC is pitched before take-off.

Both IMU's report a positive pitch (nose up) of 24 degrees from the moment they start to log ... when the AC later tries to take off the AC want to level out by lowering the pitch. If you in reality had started with the AC pointing nose up due to either placing it in a up slope or had some extra equipment on the AC that made it point 24 degrees nose up all had been fine in that leveling maneuver ... but here we see a rapid increase in the heading velocity starting directly when the AC think's it's leveling out those 24 degrees. In reality the logged up pitch of 24 degrees is very likely wrong ... during the leveling the AC instead pitches down creating that increase in heading velocity.

All seen here in the chart where I've plotted both IMU's pitch angle & heading velocity ...



At 2,5sec this is seen in the DAT log event stream ...


Can't say why the pitch was reported to be +24 degrees immediately at start up ... but this had nothing to do with magnetic disturbance. This was not a pilot error ...

 

[marcorasi1960]

After looking into the logs it's easy to understand why the AC set off forward ... Something is clearly wrong with how both the IMU's think's the AC is pitched before take-off.

Both IMU's report a positive pitch (nose up) of 24 degrees from the moment they start to log ... when the AC later tries to take off the AC want to level out by lowering the pitch. If you in reality had started with the AC pointing nose up due to either placing it in a up slope or had some extra equipment on the AC that made it point 24 degrees nose up all had been fine in that leveling maneuver ... but here we see a rapid increase in the heading velocity starting directly when the AC think's it's leveling out those 24 degrees. In reality the logged up pitch of 24 degrees is very likely wrong ... during the leveling the AC instead pitches down creating that increase in heading velocity.

All seen here in the chart where I've plotted both IMU's pitch angle & heading velocity ...

View attachment 131161

At 2,5sec this is seen in the DAT log event stream ...

View attachment 131162
Can't say why the pitch was reported to be +24 degrees immediately at start up ... but this had nothing to do with magnetic disturbance. This was not a pilot error ...

Thank you so much for the detailed analysis! Any idea what could have caused something like this? The drone was definitely on flat ground. And most importantly for me, if I fly with a brand new Mavic 2 but the same DJI Smart Controller, is there any chance that the SC was responsible for this problem and therefore I could experience it again? Thanks!!!

 

[slup]

Thank you so much for the detailed analysis! Any idea what could have caused something like this? The drone was definitely on flat ground. And most importantly for me, if I fly with a brand new Mavic 2 but the same DJI Smart Controller, is there any chance that the SC was responsible for this problem and therefore I could experience it again? Thanks!!!

No, don't have any firm opinion on why the IMU's was reporting a positive pitch of 24 degrees as soon as they became active ... it's clearly due to a hardware malfunction. The SC had nothing to do with this ... no stick inputs besides a short throttle command was made during the short time in between the take-off attempt & crash.

 

[BudWalker]

I just re-read the Flight log retrieval and analysis guide, and realized that I should have probably loaded the DAT files. So here they are. I attached 5 of them, because I am not certain which two are the relevant ones. I believe they are the first 2, but I am not 100% sure, because the hour of the flight is clearly wrong, possibly due to the time zone difference between the US (where I usually fly) and Italy (where I am now). Thanks again!

It looks like the accelX data is the cause of the pitch up indication. This plot shows the accelerometer and pitch data from FLY074 where you did a compass calibration.

I'm assuming the M2 was placed on a level surface both before and after the calibration. If so, accelX and accelY should both be close to 0.0. But, accelX = 0.45 which causes the incorrect calculated pitch.

Interestingly, accelComposite = 1.0 which probably means that an IMU calibration was done while the accelX was incorrect. Can you try doing another IMU calibration and submit the .DAT recorded while calibrating. Are you doing the IMU calibration as per DJI instructions?

Did the M2 suffer a hard knock? Like running into something? I ask because some have speculated that the accelerometers can get "stuck" when subjected to a high acceleration event.

How is it possible that BOTH accelX data is incorrect? As I have previously speculated there is only one hardware IMU. The fact that there are two IMU record types is because two different fusion strategies are used for the same IMU data. The IMU(1) data seems to incorporate vision system data where IMU(0) does not. Just a theory.

 

[marcorasi1960]

It looks like the accelX data is the cause of the pitch up indication. This plot shows the accelerometer and pitch data from FLY074 where you did a compass calibration.
View attachment 131177
I'm assuming the M2 was placed on a level surface both before and after the calibration. If so, accelX and accelY should both be close to 0.0. But, accelX = 0.45 which causes the incorrect calculated pitch.

Interestingly, accelComposite = 1.0 which probably means that an IMU calibration was done while the accelX was incorrect. Can you try doing another IMU calibration and submit the .DAT recorded while calibrating. Are you doing the IMU calibration as per DJI instructions?

Did the M2 suffer a hard knock? Like running into something? I ask because some have speculated that the accelerometers can get "stuck" when subjected to a high acceleration event.

How is it possible that BOTH accelX data is incorrect? As I have previously speculated there is only one hardware IMU. The fact that there are two IMU record types is because two different fusion strategies are used for the same IMU data. The IMU(1) data seems to incorporate vision system data where IMU(0) does not. Just a theory.

Thanks for your input! In answer to your questions:
- I have never done an IMU calibration. When I started using the M2, I read on this forum that (1) the M2 came out of the factory already calibrated and recalibrating the IMU might actually be counterproductive and (2) If and when an IMU calibration became necessary, the M2 would prompt me for it (as it does for compass calibration). Was that bad advice? In other words, should I start periodically recalibrating my new M2? If so, is there a best practice on how frequently to do it?
- My M2 did not experience a hard knock before the last 2 flights that I described.
Your input is valued and much appreciated, but I confess I am not sure I understand if there is anything different that I should do to avoid a repetition of this problem. Thanks!

 

[BudWalker]

Thanks for your input! In answer to your questions:
- I have never done an IMU calibration. When I started using the M2, I read on this forum that (1) the M2 came out of the factory already calibrated and recalibrating the IMU might actually be counterproductive and (2) If and when an IMU calibration became necessary, the M2 would prompt me for it (as it does for compass calibration). Was that bad advice? In other words, should I start periodically recalibrating my new M2? If so, is there a best practice on how frequently to do it?
- My M2 did not experience a hard knock before the last 2 flights that I described.
Your input is valued and much appreciated, but I confess I am not sure I understand if there is anything different that I should do to avoid a repetition of this problem. Thanks!

Yes, please do an IMU calibration and post the .DAT recorded during the calibration.