120ccpm Posted at 1-31 12:37
Not to intervene in the lovely debate between you, Sigmo and Labroides, but the fact that elevation values are stored in the logs (as they are) doesn't necessarily mean that the AC is using these values to calculate its height.

I think what Labroides is saying, is that the barometer is only used to translate the difference in pressure between its reading at takeoff point and current position, into a distance (height).

Not what op was asking but thought these GE pics might help regarding RTH height. White line shows terrain shape. RTH well above terrain. Possible bird strike but most likely hardware fault.

JJB* Posted at 1-31 11:51
Hi Jonny,

Franky`s MM was flying at 31 meters when loosing contact, so MM did not have to ascend as the RTH setting was 30 meters.

Hi JJB* ... ah ok ... then the RTH height was changed from 70m to 30m during flight *OMG*. Very unusual and which was definitely a very bad idea.

Pleomax Posted at 1-31 12:58
Not what op was asking but thought these GE pics might help regarding RTH height. White line shows terrain shape. RTH well above terrain. Possible bird strike but most likely hardware fault.

Well ... the only thing we can do beside speculate is to look into the provided  TXT & DAT log from the flight as there we have the facts ...

Regarding HP setting(s) & height above HP from the logs ... the blue field in the end is GoHome state, also added the TiltInclination signal to show when the impact occurred, it's in the end were the graph goes totally haywire.

The HP was initially set at 70m but were changed to 37m at second 57 & the further lowered to 30m at second 60.

The complete flight were conducted at 31m ... just above set RTH height.



If we then look at the sea level altitudes at home point & crash point ... the difference between is 11m with crash point higher. So if it's trees at the presumed crash point ... is it likely that they are 31m - 11m = 20m ? Yeah why not, very likely.



Then to confirm the above & get a grip about the terrain ...



Then someone perhaps think it wasn't so good after all to lower that RTH altitude ... yeah, well, look at the winds in the flight. They were pretty close to the maximum specified (8m/s) during the whole flight. If the original height at 70m had been kept the Mini had definitely ended up in much heavier winds at that altitude & perhaps been facing a BlowAway instead.

slup Posted at 1-31 05:17
The explanation for the position shown in "Find my drone" lays in the difference between the 2 log types, TXT & DAT.

The TXT log is generated from motor on to off ... but the DAT log is from power on to off (off state also occurs if the connection with the RC is broken).

First of all... your RTH height were set a bit to low, you clipped the tree on the highest point & crashed.

Check that again.  His drone was 115 metres above the terrain that it "crashed" into.
See this post https://forum.dji.com/forum.php? ... 059&pid=2075363    to see a profile of the terrain.

jonny007 Posted at 1-31 09:07
I'm pretty sure RTH was actually high enough. As can be seen in the first picture in post # 63, the height level of start and impact is roughly the same. The RTH altitude would have been enough to fly 50 meters above the trees. But I guess that after the connection was lost, the drone was in a very unfavorable position and before it could have ascended to RTH altitude it had previously flown into the rock or tree. Probably it was just before ascending to RTH. Therefore surely no drone error. Only the last "GPS ghost position", probably caused by a faulty GPS module due to the crash, is strange. An issue/bug (?) that does not cause any damage, although it is of course annoying that you can not rely on it and instead of search at the right place you search on the moon.

Sächsiche Schweiz, genial, da war ich vor zig Jahren auch mal, aber da gabs noch lang keine Drohnen. Auf jeden Fall muss ich da auch mal fliegen, aber natürlich nicht ganz soooo wie du ;-)

I'm pretty sure RTH was actually high enough. .. The RTH altitude would have been enough to fly 50 meters above the trees.

That's correct.
As shown in this post https://forum.dji.com/forum.php? ... 059&pid=2075363, the drone was 115 metres above the terrain we are being told it crashed into.

frankymusik Posted at 1-31 09:19
@jonny007

No it was not!

Not having set the RTH level high enough is clearly my fault. Unfortunately, nothing can be changed about this!

Hello ... there was no problem with your RTH height.
It was 115 metres higher than the terrain you are being told it crashed into.
See a profile of the terrain here:  https://forum.dji.com/forum.php? ... 059&pid=2075363
Don't be mislead by a guy that can't tell how high your drone was.

Labroides Posted at 1-31 14:31
First of all... your RTH height were set a bit to low, you clipped the tree on the highest point & crashed.

Check that again.  His drone was 115 metres above the terrain that it "crashed" into.

May I ask where you get those 115 meters from ?

slup Posted at 1-31 14:23
Well ... the only thing we can do beside speculate is to look into the provided  TXT & DAT log from the flight as there we have the facts ...

Regarding HP setting(s) & height above HP from the logs ... the blue field in the end is GoHome state, also added the TiltInclination signal to show when the impact occurred, it's in the end were the graph goes totally haywire.

If we then look at the sea level altitudes at home point & crash point ...
That's a good idea.
It might help  put this nonsense to bed.
But there's a problem with the altitude you've got for your presumed crash point.
I'm not sure how you got your altitude reading for that point, but it's nothing like what Google Earth shows.

The location to check is 50.9621  14.0716

Check it again.

The drone could not have crashed into this point because it was 115 metres above it.

slup Posted at 1-31 14:45
May I ask where you get those 115 meters from ?

May I ask where you got 292 metres from for the presumed crash altitude?

I explained 115 metres back in this post   https://forum.dji.com/forum.php? ... 059&pid=2075363

I come to this very late and don’t have the time to go through all that has been posted, but it’s only when you look at the terrain, the position of launch and the only thing the OP did on his journey was to increase throttle and move forward , if you just take a look at the map he launches way above the terrain he’s flying in, in order for him to crash he would have to lower the craft a long way. If we just look at post 81 and compare to map then it would be impossible for him to hit anything, yes he could have clipped a bird or had some kind of malfunction.

Props are held on this craft by are held on by a very small screw, this was a very new craft, I wonder did the OP bother to check screws were tight enough, if not vibration would easily loosen these and if this happened as it's quite possible then I would expect craft to react as it did, on the other hand a tiny drone crashing into a tree at 28kph, well I would expect it to be stuck in the tree, not to bounce off the tree and go backwards almost 200ft while loosing altitude .

I look at other posts tomorrow, but I think looking at the map is enough here, incidentally Rth was turned off and craft was in pgps mode no stick movement and was being moved back at end of flight .

Labroides Posted at 1-31 15:13
May I ask where you got 292 metres from for the presumed crash altitude?

I explained 115 metres back in post #42

Sorry can't see anything about 115 meters in post #42 ...

slup Posted at 1-31 16:00
Sorry can't see anything about 115 meters in post #42 ...

Sorry ... the post numbers keep changing

try this:
https://forum.dji.com/forum.php? ... =208059&pid=2075363
It should be obvious.
Look at the height of the ground below the drone   https://forum.dji.com/forum.php? ... 059&pid=2076366

Llook at how high the drone was flying

Labroides Posted at 1-31 16:05
Sorry ... the post number has changed ... try this:
https://forum.dji.com/forum.php?mod=redirect&goto=findpost&ptid=208059&pid=2075363

It's alot that seems to be to obvious to you for explaining...

Well the OP is satisfied at least ... fight on Labroides, must be hard life for you with all less educated to deal with.

Nighty, nighty

JJB* Posted at 1-31 00:57
Hi all users who can read DAT files....

Just some data recorded by this MM in the flightrecord and corresponding DAT file.

You actually saying drone crashed at 4.41, surely if drone crashed let’s say into a tree, how do you explain it fell over 100ft straight down keeping correct altitude reading , surely some common sense would tell you it would get snagged up in such a big tree, if it hit a cliff I still don’t believe mini could get to the ground in less than 4 seconds, and the most amazing thing about this crash was craft never veered off line never tumbled straight down like a bullet , so crashing into a tree is highly unlikely, and then lay upside down changed into P mode and still seemed to be moving . This was some kind of malfunction, I will look at the rest of posts tomorrow, but having looked at maps and a quick look at log, it’s one of those you can almost see what couldn’t have happened .

slup Posted at 1-31 16:13
It's alot that seems to be to obvious to you for explaining...

Well the OP is satisfied at least ... fight on Labroides, must be hard life for you with all less educated to deal with.

It is obvious ...look at this post https://forum.dji.com/forum.php? ... 059&pid=2076366     and think about it
Did you even look at this post https://forum.dji.com/forum.php? ... 059&pid=2075363

Well the OP is satisfied at least
The OP and a lot of other people here have been mislead.
Again .. look at this post https://forum.dji.com/forum.php? ... 059&pid=2076366    and think about it

slup Posted at 1-31 14:23
Well ... the only thing we can do beside speculate is to look into the provided  TXT & DAT log from the flight as there we have the facts ...

Regarding HP setting(s) & height above HP from the logs ... the blue field in the end is GoHome state, also added the TiltInclination signal to show when the impact occurred, it's in the end were the graph goes totally haywire.

Not to sure what the wind has to do with it, it certainly looked like he was flying with the wind on his tail recording 8m/s, so altitude makes not a blind bit of difference here.
I will ask you the same question I asked Jjb if drone crashed into a tree how is it that it shows no motor obstruction “as we see in almost all collisions” it then proceeds to stay in Rth while lowering a 100ft in 3 seconds and although it may look like tumbling if this was the case I think we would see a change in altitude like at least one minus number on the way down , it then changes to P MODE , the drone completely avoided the tree it just crashed into on the way down , it sounds like some kind of a pinball crash, almost like it bounced off the tree, something I have never seen before.

I think it’s more likely this craft lost or broke a prop or more like half a prop, remember these props are held on by a short screw and if not tighten they will see them selves loose , if it lost half a prop this would account for the drop straight down and the drone moving backwards almost 200ft while descending.

I have never seen a crash like this.
But drone can easily lose half a prop by screw coming loose through vibration causing  craft to spin then lose altitude while being pushed backwards.

The graphs make no sense and don’t show any impact as some are to quick to say and believe, yes they go a little crazy just as the would with loss of prop, I also have never heard of a drone impacting with a tree and immediately fly backwards and down without hitting same tree on the way down, unless tree was growing upside down ;+):::::

slup Posted at 1-31 14:23
Well ... the only thing we can do beside speculate is to look into the provided  TXT & DAT log from the flight as there we have the facts ...

Regarding HP setting(s) & height above HP from the logs ... the blue field in the end is GoHome state, also added the TiltInclination signal to show when the impact occurred, it's in the end were the graph goes totally haywire.

@slup

What should I say? Every word, every statement, every analysis result is true! Really really good. As if you were there, as an observer!
Unbelievable how exactly you can interpret the data provided!  ***)

I have never denied my guilt for the under-chosen RTH. I know it was too low!

But, and for me this is the most basic finding:
You can't really believe the location of the "Find My Drone" function!

Unfortunately, you are not able to analyze everything on site, unless you have a laptop with you and the peace and quiet of clarifying everything on site.

My conclusion:
Yes, DJI's software is also partly responsible for the fact that I could not find the drone and was therefore not allowed to use the "Care Refresh" ...
That's too bad!!!

Frank

-----------------------------
***)
The cause of the RTH (automatically triggered) was of course the broken radio connection. And then the trees were on the direct way back ...
Yes, my fault! 10m higher, and everything would have gone well ...

frankymusik Posted at 2-1 00:44
@slup

What should I say? Every word, every statement, every analysis result is true! Really really good. As if you were there, as an observer!

I have never denied my guilt for the under-chosen RTH. I know it was too low!
...
The cause of the RTH (automatically triggered) was of course the broken  radio connection. And then the trees were on the direct way back ...
Yes, my fault! 10m higher, and everything would have gone well ...
It wasn't your fault.
There was no crash because there was nothing to crash into.


Why do you ignore me when I tell you that your drone was 115 metres above the land that these idiots are telling you it crashed into?
Why do you believe them?

They are wrong.
They can't read data.
They have no idea.
Just look at this post that proves it
https://forum.dji.com/forum.php? ... =208059&pid=2075363

Or this one:
https://forum.dji.com/forum.php? ... =208059&pid=2076366

frankymusik Posted at 2-1 00:44
@slup

What should I say? Every word, every statement, every analysis result is true! Really really good. As if you were there, as an observer!

I dont think that DJI Software is even partially responsable.
The GPS Chip has reported this Position thats all the Software know.
So you can also say the GPS Chip is the Problem.
Or the GPS System itself wich had not enougth Sattelites for your special case.

The Software COULD act smarter but you could also be smart enougth to know that the drone not can fly 15km in 3 secs.

frankymusik Posted at 2-1 00:44
@slup

What should I say? Every word, every statement, every analysis result is true! Really really good. As if you were there, as an observer!

What do you think happens when you crash into a tree traveling at 28kph ?


First thing your craft stops because tree will stop it, second your log will report motor obstruction, third your drone will either get stuck in the tree or tumble through the tree.

Your drone hit the tree at 28kph and immediately bounced back off the tree and proceeded to lower altitude fast as well as moving backwards almost 200ft.

I think Laborids has clearly shown that your starting point plus the height you climbed to were far higher than the supposedly crash point, making it impossible for your craft to hit a tree.
I also think you need to look and take notice of post 81 which also shows craft was way above the height it needed to clear of any crash.

This is not the first time we have seen this , unexplained cause of craft coming down, most get put down to bird strike, but I think that from the reaction of the drone it lost a prop or in minis case half a prop, which is quite possible, but one thing I’m sure of this drone didn’t hit a tree and bounce backwards at 28kph .

Labroides Posted at 2-1 01:27
I have never denied my guilt for the under-chosen RTH. I know it was too low!
...
The cause of the RTH (automatically triggered) was of course the broken  radio connection. And then the trees were on the direct way back ...

Really all but you are idiots ?

hallmark007 Posted at 1-31 16:15
You actually saying drone crashed at 4.41, surely if drone crashed let’s say into a tree, how do you explain it fell over 100ft straight down keeping correct altitude reading , surely some common sense would tell you it would get snagged up in such a big tree, if it hit a cliff I still don’t believe mini could get to the ground in less than 4 seconds, and the most amazing thing about this crash was craft never veered off line never tumbled straight down like a bullet , so crashing into a tree is highly unlikely, and then lay upside down changed into P mode and still seemed to be moving . This was some kind of malfunction, I will look at the rest of posts tomorrow, but having looked at maps and a quick look at log, it’s one of those you can almost see what couldn’t have happened .

Hi hallmark,

Nice to see you joining in.

Crash at 4m41s. See the sudden roll angle to 167 degrees right, pitch 29 to 69 up than into negative numbers.Lookts like it hit something. (tree, rock, bird?)
After that spinning down, vertical speed down up to -20 meter per second in the first 4 seconds than up to 41 meter/second falling.
Baro height at impact 30.8 meters, dropping rapidly to 1.5 meter indicated after 3.8 second of falling.

During the fall mode remains in GoHome, apparantly no software rules to kill that mode when free-falling.
At the end of the flightlog, at minus 52 meter a general warning motorblock.
During the fall GPS position erorrs, logged GPS location data not trustfully.

Your txt "how do you explain it fell over 100ft straight down keeping correct altitude reading" ; answer above.

In the DAT file you could see the RMP values of the 4 rotors, a loosing rotor is seen rapid increase in rpm, wich is not the case,
So i rule out a loosing blade. Never read a story about MM loosing a blade because of those little tiny screws getting loose in flight.

I think some blades were severly damaged by the impact, and MM just falling and spinning down backwards.

About the height discussing. If it was flying about 115 meters above ground level at impact that point is 115-31 meters lower than HP.
AS told by others ; impact point is 11 meters higherr than HP. I did not checked the height lines in the local maps there, others did and no reason not to belief that data.

If you see the first picture in post #63, than it looks like it just fallen down along that 'clif'.

Guess we we will never know if the drone was attacked or flown against a big bird or flew against top of a tree / rock.
I simply cannot understand, as it is not explained by Labroided, to belief that this MM was flying at 115 meter above ground level that position.

Mayby you can asses and make it the 115 meter clear to everybody ?

BTW such a great place to fly, and not a single picture of video taken.

cheers
JJB

JJB* Posted at 2-1 03:24
Hi hallmark,

Nice to see you joining in.

Guess we we will never know if the drone was attacked or flown against a big bird or flew against top of a tree / rock.
You'll never know if you never look at the clear explanation I've provided.

I simply cannot understand, as it is not explained by Labroided, to  belief that this MM was flying at 115 meter above ground level that  position.
Mayby you can asses and make it the 115 meter clear to everybody ?
It's obvious to anyone that looks at the easily available information that it couldn't have hit any terrain or tree.
Take off your blindfold and look at:
https://forum.dji.com/forum.php?mod=redirect&goto=findpost&ptid=208059&pid=2076366
Check it yourself in Google Earth since you don't believe me.

Christian U Posted at 2-1 03:18
Really all but you are idiots ?

[view_image]

Really all but you are idiots ?
When I've gone to this much trouble to point out what's glaringly obvious, people that continue to act like I have no idea are idiots.
Just look for yourself at the real heights instead of the nonsense that's being pushed here.
https://forum.dji.com/forum.php?mod=redirect&goto=findpost&ptid=208059&pid=2076366

Christian U Posted at 2-1 02:16
I dont think that DJI Software is even partially responsable.
The GPS Chip has reported this Position thats all the Software know.
So you can also say the GPS Chip is the Problem.

@Christian U

... do not always repeat the same false statement, please!

Nobody else spoke of 15km. It is 1,400m between the place of the crash and the alleged last position!

You don't know anything about a 3 sec time difference (or 7 as slup wrote, see #63), you also don't know anything if you only have the RC and the smartphone with you ...

So when you're on the go (on a flight), you can only trust the Fly app.
Now one has to take a closer look to see if there were enough indications (which the Fly app should evaluate) that indicate that "ghost points" should not be used to offer them to the user as "last position".

The whole system (MM, RC, Smartphne) is full of AI, so it would be important to give the user the best possible support ...!

Frank

frankymusik Posted at 1-31 08:18
Do you think that DJI is also jointly responsible in the present case?

I don’t think anything can be held responsible for incorrect data supplied *after* a crash, I would say the “find my drone” functionality is a “best effort” type of feature which only works if the crash was not severe enough to damage the components required for this.

Labroides Posted at 2-1 03:32
Guess we we will never know if the drone was attacked or flown against a big bird or flew against top of a tree / rock.
You'll never know if you never look at the clear explanation I've provided.

Well, many others have the same question, so in your mind were are all wearing blindfolds  ;-)
You did not explain the 115 meter, just stated it and that`s it.
See all the other GE screen shots and height line charts.

But as said:  (115-31) not equal (HP - 11), and ofcourse as usual only more unfriendly words in your posting.Guess you did not checked the book i did refer too.

I will be the first to say that i am wrong, if i am wrong. Do you have the same gentleman attitude?

cheers
JJB

JJB* Posted at 2-1 03:24
Hi hallmark,

Nice to see you joining in.

I think you are not interested in looking at the terrain or indeed exactly how the craft could hit a tree at 28kph No motor obstructed and basically bounced backwards immediately, that is not anything like a crash into a tree and if little mini hit a rock going forward at 28kph it would almost certainly smash up, I think what we might be seeing here is a refusal to look at things that don’t suit the analysis, you said yourself you couldn’t be bothered checking certain things but trusting to some people that they were correct, I have looked briefly at heights at take off height climbed by the OP and height of supposed crash site and they don’t seem to match.


I know the OP was looking to find out why find my drone didn’t work out, but in all honesty I don’t think he ever knew or could see what he was doing and my theory of loosing a prop due to lack of care matches up well to how we see the craft reacting, Mavic mini is tiny it would simply get swallowed by a tree and smashed up by rocks, but it held together supposedly bounced of whatever it crashed into and flew backwards 200ft and this can be seen on playback on PH log , but no one can explain this .

I think they’re is to much emphasis on graphs and no reasoning on what we can see, and.

Labroides Posted at 2-1 01:27

that these idiots are telling you it

... do you seriously not notice that you are the only one here insulting people over and over again?..... I don’t know enough about log reading to get into the dispute between you and JJB but your attitude, man... really not cool.

hallmark007 Posted at 2-1 03:52
I think you are not interested in looking at the terrain or indeed exactly how the craft could hit a tree at 28kph No motor obstructed and basically bounced backwards immediately, that is not anything like a crash into a tree and if little mini hit a rock going forward at 28kph it would almost certainly smash up, I think what we might be seeing here is a refusal to look at things that don’t suit the analysis, you said yourself you couldn’t be bothered checking certain things but trusting to some people that they were correct, I have looked briefly at heights at take off height climbed by the OP and height of supposed crash site and they don’t seem to match.

I had expect a more serious reply from you.

I am always interested in other opions etc, just explain things to learn from each other.
Motor was obstructed few seconds later.
Saw few video where MM survived a crash into a rock **, bounded back and up. But if rotors are damaged guess no lift enougt to bounce up, gravity wins.Bouncing back with spinning damage rotors, why not falling backwards?  or bounce bounce again..

About trusting people, yes  that is a great thing to do in live. On here seeing charts from GE of height line charts from others,  yes do believe them.

You checked heights, don`t seem to match.  That`s easy saying i join in with Labroides, no explantion!
Some people on this forum don`t like to use facts and arguments, and for sure never reply om facts/arguments uses by others.

Too much emphasis on graphs, ofcourse not as showing graphs is a way of presenting and interpreting recorded data.

** btw  i flew my MA once, flying fwd with speed, against a high tension cable. What happend?  it bounced back, made a 360 turn and kept on flying (thank god)Rotors were not damaged, drone inpact area just above the cam.
So its possible to bounce back from an obstacle, for sure if that obstacle is bit flexible. (so tree top, bird mayby)

cheers
JJB

JJB* Posted at 2-1 03:52
Well, many others have the same question, so in your mind were are all wearing blindfolds  ;-)
You did not explain the 115 meter, just stated it and that`s it.
See all the other GE screen shots and height line charts.

Well, many others have the same question, so in your mind were are all wearing blindfolds  ;-)
Actually a couple of others have indicated that they believe the height was much higher than the terrain but you've failed to notice that.

You did not explain the 115 meter, just stated it and that`s it.
See all the other GE screen shots and height line charts.
But as said:  (115-31) not equal (HP - 11), and ofcourse as usual only  more unfriendly words in your posting.
I explained it again and again.
I spent hours making illustrations to make it easier for someone that can't understand simple language .

I find it very hard to believe that someone that puts himself up as someone that can interpret flight data, is incapable of:
1.  Calculating heights for himself  .. and
2.  Understanding when the details are pointed out repeatedly and in a variety of ways.

But as you seem to have no idea at all when it comes to something so basic, I'll try one more time in the most simple way.
See if this is too complicated for you.

First, look at the heights of the launch spot and the incident site.
Calculating a couple of heights is very, very basic.
Go to Google Earth and look at the flight path.
To make this easy for you, just open this Google Earth file:
http://www.phantomhelp.com/LogVi ... tRecord_2020-01-06_[11-08-43].kmz
Then hold your cursor above the home point and look toward the bottom right of your screen and find the elevation and write it down.

Then move your cursor to the incident location and check the elevation here .. write that down underneath the first one.
Then subtract the second height from the first and it should show you the difference in height between the home point and the ground level at the incident site.
Then add 31 metres because the drone was flying 31 metres higher than the home point.
Now ... did the drone run into the terrain?


Because I don't trust you to do such a simple thing, I've made some screenshots to help you.

JJB* Posted at 2-1 03:24
Hi hallmark,

Nice to see you joining in.

@JJB wrote: 'BTW such a great place to fly, and not a single picture of video taken.'

Oh my God! There must be a wonderful SD card in the MM, which allows breathtaking insights into this wonderful nature.

Unfortunately, the "real" flight time was at most 2 minutes, but of course it should actually take about 20 minutes ...
So sad.

Frank

frankymusik Posted at 2-1 03:44
@Christian U

... do not always repeat the same false statement, please!

Ok, es ist ein bisschen übertrieben, aber nicht unwahr.

The 1,4km would need 1,8 MINS at maximun Speed.
And jeah i can understand that, you was nervous and have lost your drone in that moment.

BUT its very nessesary to keep cool during flights. You fly clearly out of sight and over Houses wich is not allowed in germany. This roules exist becouse people dont think sometimes during flying. So i think we have especially in Germany a lot of rules and its sometimes ok to not get used roo bad into them. But then you have to use your head and calm down during flight.

Yeah ... every crash analyze start up with a stack of possible scenarios (usually several as we weren't there when it happened). Then it's time to seek support from the logs to circle in the most possible cause as it's there the facts are stored ... without looking at those everything just remains speculative ... looking at graphs generated directly from the logs are one way to to see the patterns, others may instead prefer to look at the values directly in a csv reader or in a web based service like Airdata.com or PhantomHelp. In the end of every crash investigation some lose ends will remain ... but coming that far you usually can point out the most probable cause.

It's always important to be open minded & not lock in to one scenario to early ...

So let's see what theory the logs can support ... a lost prop or an impact into foliage, what is more likely.

First a bit about the error & info messages in the log:

First we can see that the GPS module restarts (likely here the wrong position is created), then alot of gimbal motor errors occurs usually occurring when something prohibit the free movement of the gimbal . And in the end we see that the motors no longer can rotate.



Then a look at the prop rpm's ... if a prop is lost that engine would immediately max out. But that's not shown in the log. Instead all 4 props get lowered rpm's in four distinct periods until all go to zero in the end. Nothing indicate that one of them ... or actually none of them have been lost, more that they are under influence from something alot denser than air but not continuously though as the rpm's of all 4 gets intermittent.



So what about the motor current draw ... again pretty much 4 distinct periods were the current draw especially for the 2 front props rise but without going over the top ... a bit like that they chews into something softer. But nothing here either that indicates that a prop has been lost.



And then lastly a bit from the logs about acceleration directions when the event starts ...
(+Z is down, +Y is horizontal right & +X is horizontal forward). The DAT log indicate a rapid movement down & backwards but nearly no change left/right... like a sudden resistance coming from above front. The values in the logs don't show a sudden stop (like if you've made contact with a cliff wall) instead it again show resistance in 3 distinct periods at + altitude as if the Mini tries to escape a foliage.

Ice_2k Posted at 2-1 03:58
... do you seriously not notice that you are the only one here insulting people over and over again?..... I don’t know enough about log reading to get into the dispute between you and JJB but your attitude, man... really not cool.

... do you seriously not notice that you are the only one here insulting people over and over again?
I've been analysing flight data and helping drone flyers all over the world for four years .
Usually people thank me for the effort I put in.
I have never had so much trouble getting basic concepts through to anyone, let alone someone that acts like he's an expert.
I've tried again and again to explain but all he does is tell others that I don't know what I'm talking about.

Do you seriously think that I haven't double (and triple) checked my working to see if that idiot with the graphs might be right?

..... I don’t know enough about log reading to get into the dispute between you and JJB but your attitude, man... really not cool.
Most of the people here are in the same position, but that doesn't stop them from assuming that I have no idea about data analysis.
See if you can see what I've been trying to explain in my previous post .. the one with the pictures.

Ice_2k Posted at 2-1 03:50
I don’t think anything can be held responsible for incorrect data supplied *after* a crash, I would say the “find my drone” functionality is a “best effort” type of feature which only works if the crash was not severe enough to damage the components required for this.

@ice_2k

Of course, that's a point ...

But, maybe it will be possible to sort the events "automatically" when the software "knows" that the motors have stopped, the GPS signal is no longer available ...
Then the software should rather output a position, which is in the vicinity, but not 1,400m away .......

f not everything broke, you could still use the "beep and blink" function to find the MM ... (... but this cannot work at a distance of 1,400m).

Frank

Labroides Posted at 2-1 04:38
Well, many others have the same question, so in your mind were are all wearing blindfolds  ;-)
Actually a couple of others have indicated that they believe the height was much higher than the terrain but you've failed to notice that.

Thanks,

Now you are explaining!  

I hope i can understand it...hahaha
If you just write without making silly remarks i might start to like you.

Think i see now what is wrong, your point is bit to the S. so elevation is along that clif to lower point.
As said by others, that area there is 11 meters higher than HP. but just before that ofcourse lower.

As data in the log is little delayed written, drone can be just a little more to HP.
See if i can make some calculations on latlong versus time using fwd speed.

cheers
JJB

Labroides Posted at 2-1 04:46
... do you seriously not notice that you are the only one here insulting people over and over again?
I've been analysing flight data and helping drone flyers all over the world for four years .
Usually people thank me for the effort I put in.

Lovely graphs .. but you still haven't bothered to look at the heights involved.

JJB* Posted at 2-1 04:23
I had expect a more serious reply from you.

I am always interested in other opions etc, just explain things to learn from each other.

I think I know now I’m right, your not prepared to move one inch from your position, and it’s clear that you will only accept information that only suits your argument.
Your graph doesn’t tell you of any crash it just tells you something happened nothing else, 95% of people just look at the graphs as though they are colored pictures and want an explanation.

You use arguments like a high tension wire bounced your Mavic air, yes if the mini hit high tension wire it might bounce, but your claiming first it hit a tree then you claim it hit a rock, in both of those situations I wouldn’t expect mini to bounce back some 200ft in fact I would expect into rock at 28kph to be destroyed and into a tree I’d expect it to be in the tree or fall through the tree but to bounce of a tree and come back 200ft is beyond belief.

And there is no motor obstruction on impact , but on almost every impact crash it’s the first telltale warning we see show up.

You seem to be able to make up your own stories all props were damaged, I mean did you just make this up or how do you know this, your fond I telling others that when wrong you step up to the plate, I’d have to say in my experience around here this is far from the truth .

I think the technical arguments have been made, and from what I see your being a bit pigheaded only choosing to look at what suits your argument which is wrong .

slup Posted at 2-1 04:45
Yeah ... every crash analyze start up with a stack of possible scenarios (usually several as we weren't there when it happened). Then it's time to seek support from the logs to circle in the most possible cause as it's there the facts are stored ... without looking at those everything just remains speculative ... looking at graphs generated directly from the logs are one way to to see the patterns, others may instead prefer to look at the values directly in a csv reader. In the end of every crash investigation some lose ends will remain ... but coming that far you usually can point out the most probable cause.

It's always important to be open minded & not lock in to one scenario to early ...

@slup and of course to everyone else:

What a brilliant interpretation of the incident!
Man, how much time you take for this matter !!!
I can only thank you again and again for the much knowledge that you have imparted!
Other users will feel the same, thank you for that ...

I can only repeat it, your descriptions match our memory as much as possible. Terrific. Thank you.

So many people have thought about the incident here, and everyone deserves our special thanks!

For me it was really only important to find out where this position PDrone comes from? Invented or already taken from a log. It was resolved thanks to slup. That's why all of this isn't as mysterious anymore as first thought ...

Therefore again: Many thanks to everyone and especially to the one !!!


Frank