There’s no fool like an old fool...Mavic Mini lost!!

[Meta4]

I have found this. Is this what we are looking for?

Very good - thanks.
The flight looks like this: DJI Flight Log Viewer - PhantomHelp.com
I'll see what the data is saying and get back soon.

 

[OldGuy]

Very good - thanks.
The flight looks like this: DJI Flight Log Viewer - PhantomHelp.com
I'll see what the data is saying and get back soon.

Thanks for that. Now I can see where it really went. I thought it was higher up on the side of the mountain. I wish I had this information before perhaps I could have located it. In any case I will go over there again and see if anyone has found it.

 

[Dan C]

Please read my answer to Meta4 which completes his analogy by the relation of a little personnal anecdote.
Maybe you'll realize that a new calibration in a perturbated environment can lead to catastrophic events when the perturbation diminishes, changes or disappears completely (my cans which are used one by one for cooking).
For example if you are with your drone close to a loudspeaker, it is very likely that the app will ask for a compass calibration. Do it for "peace of mind" and I guess that your peace will vanish very rapidly just after take off

That’s incorrect.

If you read sar104’s post about what calibration does, you’d realize that calibration only identifies the fields that don’t change as the drone is rotated.

Your analogy fails because the cans’ position is static relative to the boats compass.

If your drone, on the other hand, is affected by a large ferrous object nearby, a compass calibration will correctly exclude that influence.

It will likely not show magnetic north correctly, but that would resolve once you fly away from the object.

 

[Meta4]

Now here is my first mistake I didn’t do a compass calibration which is most unusual for me as I am a big believer in CC.

The drone flew properly, could hold a straight course and hover in place without slowly spinning.
There was no problem with its compass (which is not surprising).

The next mistake was not hearing “The Home Point has been updated please check it on the map” message.

The data shows that the drone had perfect GPS reception right from the start and recorded its home point just as is normal.

I didn’t wait for the RTH to kick in. I reasoned that if I pulled back on the remote stick I would fly backwards along the outgoing flight path.

You flew out at a height of 94 ft on a heading of 220° at full stick and a speed of 18 mph until you took your hand off the right stick at 5:08.
At 5:12.2 you pulled the right stick back full.
The drone came back on a slightly different course of 20° at 18 mph until 6:48.3 when the data stops with the drone still up 95 feet and 1220 ft from the home point.

A look at the terrain shows that the point where the track ends is 92 feet higher than the launch point (according to the approximate heights in Google Earth)

It looks very much like the drone will have bumped into a house, wall etc over there at 36.59428 -4.63605.

It turns out that your assumed mistakes weren't the problem.
The drone coming back on a slightly different course took it toward higher ground.
It would have been impossible to plan for that but a higher RTH height would have prevented the collision.

You still had signal as shown by the data and flight details after losing your app view.
If you had switched off the controller to lose signal, the drone would have initiated RTH.
There doesn't appear to have been any wind problem.

​

​

 

[Meta4]

If your drone, on the other hand, is affected by a large ferrous object nearby, a compass calibration will correctly exclude that influence.
It will likely not show magnetic north correctly, but that would resolve once you fly away from the object.

And if that was correct, I wonder why DJI advise:

Particularly the point marked by the lightbulb (suggesting it's important perhaps?)
But as you seem to be the ultimate expert on compass calibration, I'm sure you'd understand all about that.

Don’t calibrate around your cell phone, eh?
That would be a neat trick since my cell is the device running the software telling me to calibrate.

BTW while you are rereading DJI's recommendation above, you might note that they don't say anything about being near your phone.
They advise not to have your phone on you while calibrating.
If you put the controller and your phone down and are a metre or two away, it will have no detectable influence on the compass you are calibrating.

 

[Dan C]

And if that was correct, I wonder why DJI advise:

Particularly the point marked by the lightbulb (suggesting it's important perhaps?)
But as you seem to be the ultimate expert on compass calibration, I'm sure you'd understand all about that.

BTW while you are rereading DJI's recommendation above, you might note that they don't say anything about being near your phone.
They advise not to have your phone on you while calibrating.
If you put the controller and your phone down and are a metre or two away, it will have no detectable influence on the compass you are calibrating.

Well, you did after all previously state:

As has been explained several times in this thread, calibrating the compass only identifies and measures the magnetic fields that are part of your drone.

So which is it? What actually is your position?

Or are you simply going to gainsay anything I post?

 

[deathmajor]

My lessons from reading this thread:

1) He should have initiated RTH with the button on his controller and made sure it had appropriate height set.
2) The drone can alter course even if you don't touch the left stick.
3) Don't break the law, fly within VLOS.

Also, seeing the flight log, it's a reminder that the worst thing about a drone going missing - isn't the loss of a drone. It's the fact that it comes crashing down and might cause damages. I saw a drone YouTuber today who flew on top of a car bridge in dusk above water. After flying into the bridge and crashing, he blamed the drone for not avoiding the obstacle. Imagine if he had hit a car or just startled the driver to steer into the railing.

 

[OldGuy]

The drone flew properly, could hold a straight course and hover in place without slowly spinning.
There was no problem with its compass (which is not surprising).

The data shows that the drone had perfect GPS reception right from the start and recorded its home point just as is normal.

You flew out at a height of 94 ft on a heading of 220° at full stick and a speed of 18 mph until you took your hand off the right stick at 5:08.
At 5:12.2 you pulled the right stick back full.
The drone came back on a slightly different course of 20° at 18 mph until 6:48.3 when the data stops with the drone still up 95 feet and 1220 ft from the home point.

A look at the terrain shows that the point where the track ends is 92 feet higher than the launch point (according to the approximate heights in Google Earth)

It looks very much like the drone will have bumped into a house, wall etc over there at 36.59428 -4.63605.

It turns out that your assumed mistakes weren't the problem.
The drone coming back on a slightly different course took it toward higher ground.
It would have been impossible to plan for that but a higher RTH height would have prevented the collision.

You still had signal as shown by the data and flight details after losing your app view.
If you had switched off the controller to lose signal, the drone would have initiated RTH.
There doesn't appear to have been any wind problem.

​

​

Wow what a lot of data from the flight record. I have had drones for several years but have never seen this kind of detail. I note that the Home Point was recorded which is good. So that wasn’t a problem. It looks like the change of course on the return was me pulling back on the stick. I should have left it alone?

I have just come back after visiting the villa where my MM ended it’s flight. The place is locked up for the winter with no one around. I climbed over the balustrade and accessed the pool area and then the next balcony up but found nothing. There is still another balcony exiting a bedroom which is not accessible. I looked over the roof area with binoculars but could see nothing that looked like a MM.

Tomorrow I will fly my M2Z over the place and take a look at the upper balcony and also to get a better look at the roof. If it missed the roof of the villa then it will have hit the rough rocky ground behind which is overgrown and inaccessible. We will also leave a note for the owners in case they visit the place in the near future.

I can’t thank you enough for helping me. I should have joined Mavic Pilots ages ago as everyone is very helpful and I have learned a lot from all the posts and of course will keep on learning.

 

[OldGuy]

My lessons from reading this thread:

1) He should have initiated RTH with the button on his controller and made sure it had appropriate height set.
2) The drone can alter course even if you don't touch the left stick.
3) Don't break the law, fly within VLOS.

Also, seeing the flight log, it's a reminder that the worst thing about a drone going missing - isn't the loss of a drone. It's the fact that it comes crashing down and might cause damages. I saw a drone YouTuber today who flew on top of a car bridge in dusk above water. After flying into the bridge and crashing, he blamed the drone for not avoiding the obstacle. Imagine if he had hit a car or just startled the driver to steer into the railing.

1. I did hit the RTH button but I had a blank screen on the tablet and all the lights on the remote were flashing so I had no idea what was happening. Of course with hindsight I would act differently now. I came to that conclusion a few minutes after loosing the MM. The RTH height was already set correctly. To clear this valley in all directions I would have to set the RTH at 1.3 Km. The flight path I was taking and expecting to return on was well below my RTH setting with nothing in the way.
2. How does a drone alter course if you don’t touch the sticks. You've got me on that one.
3. I’ve already said that I expect some stick (pun) from the dedicated VLOS brigade.

 

[sar104]

That’s incorrect.

If you read sar104’s post about what calibration does, you’d realize that calibration only identifies the fields that don’t change as the drone is rotated.

Your analogy fails because the cans’ position is static relative to the boats compass.

If your drone, on the other hand, is affected by a large ferrous object nearby, a compass calibration will correctly exclude that influence.

It will likely not show magnetic north correctly, but that would resolve once you fly away from the object.

No - that's still not right. The field components that don't change (in the aircraft's frame of reference, since that's the fixed frame of reference for its magnetometers) as the aircraft is rotated, are those of the magnetic field of the aircraft, not the surroundings. It cannot distinguish the magnetic field from a nearby magnetized object from the magnetic field of the earth. Both of those change together as the aircraft is rotated.

 

[Dan C]

No - that's still not right. The field components that don't change (in the aircraft's frame of reference, since that's the fixed frame of reference for its magnetometers) as the aircraft is rotated, are those of the magnetic field of the aircraft, not the surroundings. It cannot distinguish the magnetic field from a nearby magnetized object from the magnetic field of the earth. Both of those change together as the aircraft is rotated.

And I said, "...calibration only identifies the fields that don’t change as the drone is rotated." From the aircraft's frame of reference, everything revolves around it -- hence what does not change is the aircraft's field.

That is the same thing. How do you not see that? You are demonstrating a lack of conceptual flexibility here that suggest you are operating in the lower half of Bloom's Taxonomy.

Then you say:

It cannot distinguish the magnetic field from a nearby magnetized object from the magnetic field of the earth.

I never said it did. You are, once again as you have done throughout this to-and-fro, created a straw man argument to attack.

Your apparent inability to recognize a concept in a different reference frame, and your rigid adherence to a rather formulaic expression of this knowledge, are making me think you are reading this stuff from a book.

You have squandered a teaching opportunity -- if indeed you truly grasp the concepts here. At every turn your only attack has been to misrepresent or outright fabricate what I've stated, then defeat your own argument.

You have consistently implied, and Meta4 has blatantly accused me, of willful ignorance and inflexibility in this matter. But it was MY statement which you both attacked. You do not get to re-frame my position to suit your argument. But straw man arguemnts is all you've brought, and you've both been insulting in the doing of it.

It has been invigorating, and the process has clarified and strengthened my understanding of this whole calibration issue. And in the end, other than a few possible points of semantics, you have both been shadow boxing, and not actually shown anything I actually said (not what you claim I said) as wrong. For that I thank you both.

But it is clear to me that neither of you have anything further of value to offer me on this matter. And since you both have high status here, neother of you will back down or compromise. That's fine.

Seriously, though... I'm no longer interested in playing Alice to your Rabbit Hole. This just has nowhere to go except crazy town.

 

[sar104]

And I said, "...calibration only identifies the fields that don’t change as the drone is rotated." From the aircraft's frame of reference, everything revolves around it -- hence what does not change is the aircraft's field.

That is the same thing. How do you not see that? You are demonstrating a lack of conceptual flexibility here that suggest you are operating in the lower half of Bloom's Taxonomy.

Then you say:


I never said it did. You are, once again as you have done throughout this to-and-fro, created a straw man argument to attack.

Your apparent inability to recognize a concept in a different reference frame, and your rigid adherence to a rather formulaic expression of this knowledge, are making me think you are reading this stuff from a book.

You have squandered a teaching opportunity -- if indeed you truly grasp the concepts here. At every turn your only attack has been to misrepresent or outright fabricate what I've stated, then defeat your own argument.

You have consistently implied, and Meta4 has blatantly accused me, of willful ignorance and inflexibility in this matter. But it was MY statement which you both attacked. You do not get to re-frame my position to suit your argument. But straw man arguemnts is all you've brought, and you've both been insulting in the doing of it.

It has been invigorating, and the process has clarified and strengthened my understanding of this whole calibration issue. And in the end, other than a few possible points of semantics, you have both been shadow boxing, and not actually shown anything I actually said (not what you claim I said) as wrong. For that I thank you both.

But it is clear to me that neither of you have anything further of value to offer me on this matter. And since you both have high status here, neother of you will back down or compromise. That's fine.

Seriously, though... I'm no longer interested in playing Alice to your Rabbit Hole. This just has nowhere to go except crazy town.

It's unfortunate that you are still misunderstanding this subject. And in post #83 you stated exactly what you now claim not to have stated:

If your drone, on the other hand, is affected by a large ferrous object nearby, a compass calibration will correctly exclude that influence.

To which I replied:

No - that's still not right. The field components that don't change (in the aircraft's frame of reference, since that's the fixed frame of reference for its magnetometers) as the aircraft is rotated, are those of the magnetic field of the aircraft, not the surroundings. It cannot distinguish the magnetic field from a nearby magnetized object from the magnetic field of the earth. Both of those change together as the aircraft is rotated.

To which you replied:

I never said it did. You are, once again as you have done throughout this to-and-fro, created a straw man argument to attack.

That small part of this discussion really sums it up. Both @Meta4 and I have very patiently tried to explain this to you, and you have steadfastly refused to listen. It's not a teaching opportunity that has been missed on our part - it's a learning opportunity that you have missed. On top of that your attitude has simply become more combative and unreasonable as the discussion progressed, and so I think your decision to back away is the best one.

And by the way - neither @Meta4 nor myself have any special "status" on this forum - we are both just regular members.

 

[EricJT]

USER=1601]@Meta4[/USER] and I have very patiently tried to explain this to you, and you have steadfastly refused to listen. It's not a teaching opportunity that has been missed on our part - it's a learning opportunity that you have missed.

No sarcasm intended on my part, but I have found this thread to be both educational and entertaining. I now know more about my M2P and magnetic and electromagnetic theory than I did before following this thread. When I am unable to continue to learn I will more than likely be on the wrong side of the daisies.

 

[BradinLae]

"a dwarf or giant in Scandinavian folklore inhabiting caves or hills."
What an interesting world we live in. It took 5 pages but eventually we discover what really happened. There has been some good information and some bad.
Something to note is that the developers of the software are almost certainly not the same engineers that design the aircraft and therefore may have some of the misconceptions about things that users may. Of course Corporate will say we have had to replace X number of drones because of compass problems tell the software guys to fix that.
Software team will have a meeting and instead of having one message on the screen that says "move the aircraft away from the steel or magnetic field" and another for "please calibrate your compass as we believe there may have been a change in your aircraft's configuration." They have the same message pop up whenever there is an out of range reading of the sensors.

 

[GregE]

Interesting discussion I'll follow.. That was after reading page one
Glad helpful folks were able to help with the log track info.
Good to see several seasoned pilots on here

 

[Captain.rick]

I hate to be added to the list of lost Mavic Minis but here I am. However I must say that it was my fault not DJI as I panicked! I’ve been flying R/C for over 40 years and now have a few years on drones so there is no excuse. This was my second MM having returned the first one due to a disappointing CE range. So here I am on my favourite hill 18th December with nothing around for miles. I fly the FCC version of the MM out almost 2 km and I am very happy due to the excellent range that it has. Unfortunately a cold wind started to blow so I packed up and set off home. Once home as I had proven the FCC MM I decided on one more flight from my garden before calling it a day. Now here is my first mistake I didn’t do a compass calibration which is most unusual for me as I am a big believer in CC. The next mistake was not hearing “The Home Point has been updated please check it on the map” message. I took off full of confidence and flew out across the valley (my place is 350 m above sea level on the side of a mountain). Next mistake was my premature over confidence. Instead of flying VLOS (I know you guys are going to jump on me because of this) I flew out around 600 m saying to myself “I've got a FCC MM so no problem”!! Suddenly everything went dead. I lost the tablet screen, all 4 lights on the remote stared flashing. Now here is the final mistake. I didn’t wait for the RTH to kick in. I reasoned that if I pulled back on the remote stick I would fly backwards along the outgoing flight path. However I should have put the remote down and waited for the RTH.
A futile search of the are revealed nothing. Find My Drone showed the aircraft heading to the coast and out to sea. The Flight Data Centre showed the outgoing flight path and the beginning of a RTH which was interrupted (by me fiddling with the sticks I suppose) which then resulted in the aircraft flying into the side of the mountain.
I returned to my office and immediately ordered another MM CE version which on a good note arrived on Christmas Eve. Since then I have flown this one several times and by switching to manual in Channel Mode have managed to get out around 2 km on 2.4 GHz. I didn’t try this with my first MM unfortunately.
I still look out over the valley and wonder what happened to my poor ill fated Mavic Mini.
I now am the proud owner of a Mavic Mini FCC remote but no drone! By the way I can confirm that it won’t pair with my CE drone!!
Final note: I forgot to put my phone number on the aircraft….all my other drones have it!!

OldGuy/Old Fool!!

What's an FCC version and a CE version? Thanks

 

[StinkiePants]

I found this subject matter interesting, so thanks. But most of the thread was agonizing. It seems on every forum I go to, there is a lot of bickering and what I would call 'grumpy old man' attitude, which ends up causing digressions and (what I'm sure is) painful amounts of time wasted writing responses. Belittling comments and things like, "wrong", "what you don't/can't understand" should be left at home. Reminds me of being a kid and the neighbor that was staring out the window waiting to yell, "get off my lawn!" Just my opinion.

 

[bonmot]

it is not something you need to do before every flight unless the drone asks for it ,sometimes you will get calibrate compass simply because you are too near to your car or there is some metal in the ground you can't see,it is probably a routine that OLD GUY has been doing for many years because of something he read about it ,i have a routine where i use a small hand held compass to find magnetic north and set my take off mat to north before i fly and i swear by it and won't fly until i have done it that way i know that my Mavic is facing north because i can visually see it on the screen

I like the idea of a hand held compass! I should put one in my new M2Z case. Or at least check the compass on the iPhone.
And take off mat to north is another good idea!
But just wondering about the third thing: How can I check what my drone thinks the north is?
Thanks

 

[sar104]

I like the idea of a hand held compass! I should put one in my new M2Z case. Or at least check the compass on the iPhone.
And take off mat to north is another good idea!
But just wondering about the third thing: How can I check what my drone thinks the north is?
Thanks

The aircraft orientation arrow on the map display indicates the IMU heading.

 

[old man mavic]

if you have an ipad the orientation display that looks like a radar screen is on the left at the bottom and the map is on the right, in the circle around the orientation display you will see an N ,and the triangular shape in the middle of the display ,which represents the drone, will be pointing towards the N if the drone is correctly aligned