Hi folks,

I was testing my drone (Air 2S) after fixing a rear obstacle avoidance sensor issue last evening after calling customer service, and although this fake obstacle alert issue has been resolved, my drone crashed itself into a building under construction during this unfortunate flight. The incident happened after my drone's signal got disconnected at about 0.5 miles away, and the auto return home function got activated during 'Normal' flight mode. During the RTH, I can see intermittent image transmitted back to my controller and I was able to see it was moving right torward a building. I desperately commanded it to stop the RTH function, but it won't respond. No matter how many times I click the stop RTH button, the transmitted back video is still showing it is moving toward the building without hesitation. I guess the signal was too poor to carry my command to the drone. Moments afterwards, I lost connection for good. Only pitch black screen with "Account is not logged in" warning. I waited for about 3 minutes and nothing happened. It didn't came back. It crashed itself right into a building under construction. The height was about 120m so there is no way it can survive the impact, what's making it worse, it was a private ground under construction in the evening so I cannot just go in and pick it up (If I can find the debris in the first place).

Now, my questions are:
1. Why the frontal obstacle avoidance sensor totally did nothing? Was the obstacle avoidance sensors disabled during RTH function? It was under Normal mode and I can see the obstacle avoidance was in activated state. Why it is so dumb that it crashed itself during a RTH and conducted a frontal impact to a wall?

2.I do have the DJI Care Refresh (1-Year Plan) Service, which I just purchased 2 days ago after I activated my drone. But I guess it doesn't cover the case where my drone is totaled and cannot even be retrieved back, does it?

I had a traumatic experience for this second drone of mine. I had a first generation DJI Mavic Pro before and although the camera was bad, I didn't have so much issues with the flight itself.

Sorry for your loss of the drone. Hope Dji give you some satisfaction.

The front obstacle avoidance might have worked and the drone hovered in place before battery levels forced a landing -- the drone might still be recoverable if you can get on site or contact someone on site to look for it. The building structure might have completely blocked any radio communication, triggering the app behavior you report. Good luck!

This is why you set the RTH altitude HIGHER than local obstacles.

Ex Machina Posted at 12-11 11:48
The front obstacle avoidance might have worked and the drone hovered in place before battery levels forced a landing -- the drone might still be recoverable if you can get on site or contact someone on site to look for it. The building structure might have completely blocked any radio communication, triggering the app behavior you report. Good luck!

I considered that possibility, but I walked there in 10 mins and saw nothing hovering -- in theory, it has about 90% battery when I lost its signal so it should have enough battery to hover for another 10 mins...

Mobilehomer Posted at 12-11 12:51
This is why you set the RTH altitude HIGHER than local obstacles.

Lessons learnt, but my RTH altitude IS higher than that building -- It is not the first time it lost signal, but this is the first time it didn't attempt to fly to 120m before returning home. I don't know why. --Edit: Correction, after investigation, my RTH height is lower than the building. I made a mistake thinking that building was lower than my preset height.

Do you have the flight log or maybe logs (plural) from this flight, if so can you upload them to either
https://www.phantomhelp.com/LogViewer/Upload/
or
https://app.airdata.com/
and post the link or links here, so that the folks who can read them can have a look and offer more than guess work.
The former page has instructions on how to access you logs.

For care refresh you generally, if not always, need to retrieve the drone.

Is it possible that your repeated pressing of the RTH/Pause button alternated between cancelling an RTH and initiating one?
Can you check the app to see if the RTH height is set to the height you think it was?

fans63495e80 Posted at 12-11 13:08
Lessons learnt, but my RTH altitude IS higher than that building -- It is not the first time it lost signal, but this is the first time it didn't attempt to fly to 120m before returning home. I don't know why.

Flight data is always the first place to go to solve flight incident mysteries.

Go to http://www.phantomhelp.com/LogViewer/Upload/
It has instructions to find the .txt file from your phone or tablet which you can upload to that site
Upload it and it gives you a summary of the flight data.
Post a link to that summary.

Or just post the .txt file

Sean-bumble-bee Posted at 12-11 13:49
Do you have the flight log or maybe logs (plural) from this flight, if so can you upload them to either
https://www.phantomhelp.com/LogViewer/Upload/
or

Thanks for the advice! I checked DJI Fly app from my phone, I can see my last flight data under My Profile but I don't know how to access the txt log file out of the app. But here are some useful data I have retrieved: at flight time 2:33, the drone was moving at RTH altitude 119m.  Vertical speed was 0.0m/s and Horizontal speed was 12.0m/s. The next data point is captured at time 2:40 -- There are no data points between 2:33 and 2:40. Altitude was 90m, Vertical speed was -16.0m/s and Horizontal speed was 8.2m/s. Then the last data point was captured at 2:44, altitude was 9.1m, Vertical speed was -19.9m/s and Horizontal speed was 5.7m/s. Turns out the height was at RTH height 120m, the building was taller than it, at approximately 2-3 floors higher, around 135m. I thought 120m was high enough and I didn't want to break local restrictions.

Labroides Posted at 12-11 14:46
Flight data is always the first place to go to solve flight incident mysteries.

Go to http://www.phantomhelp.com/LogViewer/Upload/

Thanks for advice, do you know how can I extract flight data out of DJI Fly app? Cloud?

fans63495e80 Posted at 12-11 19:23
Thanks for advice, do you know how can I extract flight data out of DJI Fly app? Cloud?

Follow the instructions in either link that Sean and Labroides posted. Or watch this: https://app.airdata.com/wiki/How ... %80%99s+Android+App
The file is on your mobile device.

Hi there, we are sorry to hear about the accident. We recommend that you start a case through the link below first, if there is a doubt about the accident, data analysis can be applied as long as the drone is within the warranty period. http://www.dji.com/support

Sean-bumble-bee Posted at 12-11 13:49
Do you have the flight log or maybe logs (plural) from this flight, if so can you upload them to either
https://www.phantomhelp.com/LogViewer/Upload/
or

I've attached flight log here: https://www.phantomhelp.com/LogViewer/9A8S8SUWHEZUTPHAQIPWNow, from the look of ithe log, "the ambient light was too low and obstacle avoidance was not available". Is the obstacle avoidance system built on top of visibile light spectrum rather than radar? This could be the colision cause...

Labroides Posted at 12-11 14:46
Flight data is always the first place to go to solve flight incident mysteries.

Go to http://www.phantomhelp.com/LogViewer/Upload/

I cannot post txt file in forum so I posted it here: https://www.phantomhelp.com/LogViewer/9A8S8SUWHEZUTPHAQIPW
From the look of the log, "the ambient light was too low and obstacle avoidance was not available". Is the obstacle avoidance system built on top of visible light spectrum rather than radar? This could be the collision cause...

DJI Mindy Posted at 12-12 01:09
Hi there, we are sorry to hear about the accident. We recommend that you start a case through the link below first, if there is a doubt about the accident, data analysis can be applied as long as the drone is within the warranty period. http://www.dji.com/support

Thank you Mindy. I called support but right now it seems it is not in office hour. I will try raise a case and follow up, thanks again!

As far as I know the OA uses visible light and I think you have worked out the reason for the crash, sorry. I think your best hope is to try recover the drone and care refresh it.

fans63495e80 Posted at 12-12 07:55
I cannot post txt file in forum so I posted it here: https://www.phantomhelp.com/LogViewer/9A8S8SUWHEZUTPHAQIPW
From the look of the log, "the ambient light was too low and obstacle avoidance was not available". Is the obstacle avoidance system built on top of visible light spectrum rather than radar? This could be the collision cause...

Hi,

Had a look at your log.

At 2m12s you lost connection, your AIR2S into RTH. Guess the signal was blocked by high building, if your were located on the ground.
See the RTH fly path direct line to home in the chart.

Flying at 392 feet height, crashing into the buidling. I have not checked the height of the building, as the building is not on the map....
OA did not function, several warnings in the log "Ambient light too low. Vision system and obstacle sensing unavailable. Fly with caution"

OA sensors use normal light, so in the dark they do not work.
sorry for your loss.

cheers
JJB

Sean-bumble-bee Posted at 12-12 11:33
As far as I know the OA uses visible light and I think you have worked out the reason for the crash, sorry. I think your best hope is to try recover the drone and care refresh it.

That's a bummer, but at least I know why it happened. Any idea why other radar wavelength was not used? And, even visible wavelengths were used, I can't imagine how that could affect the radar. From what I know, radar does not rely on ambient light; they emit their own electromagnetic waves. Ambient light cannot be a factor for radar to dysfunction exactly because of this. This has been how radar works since WWII, the british knows air raid incoming day or night, regardless of ambient light conditions...

if you are going to fly your drone around buildings that are (much) higher than 400 feet (here in the usa where it is unlikely you will set your rth to higher), you must always be extra careful and be especially aware of what you are doing.  if you're only flying for fun, it makes no sense to fly around the other side of a massive building since that's just asking for trouble.  usually there are quite a few building at least if not more than 400 feet higher than where you are standing and unless you are willing to always maintain vlos and take it easy, even an experienced pilot is going to potentially run into problems.  i hope you get your drone back since that would make it easier to claim.

The Saint Posted at 12-12 12:04
if you are going to fly your drone around buildings that are (much) higher than 400 feet (here in the usa where it is unlikely you will set your rth to higher), you must always be extra careful and be especially aware of what you are doing.  if you're only flying for fun, it makes no sense to fly around the other side of a massive building since that's just asking for trouble.  usually there are quite a few building at least if not more than 400 feet higher than where you are standing and unless you are willing to always maintain vlos and take it easy, even an experienced pilot is going to potentially run into problems.  i hope you get your drone back since that would make it easier to claim.

Lessons learnt, in the hard way. I made a series of seemingly trivial mistakes that ultimately led to this incident. I guess it is a lesson that I will have to learn at some point, sooner or later.  I assumed the obstacle avoidance system during RTH was smart enough to not to collide into a building directly in 2021. And obviously that was not a valid assumption, it is still not quite as 'smart' as I imagined from commercials... As the radar was somehow affected by ambient visible wavelength light conditions, something I failed to comprehend even today. As a pilot with plenty of experience flying in open plain but limited experience flying around tall buildings, I have yet to have touched the boundary of how far this obstacle avoidance and RTH can save me. This is a new building under construction, so there is no telling how high it is  before flight.Multiple factors led to this incident and here I am.

JJB* Posted at 12-12 11:40
Hi,

Had a look at your log.

Thanks for the analysis, it matched everything we concluded so far. The building under construction is now around 135m/440 feet tall, just a bit higher than RTH 120 m height. Do you know why OA radar can be affected by ambient light? As far as I know, radars emits electromagnetic waves in different wavelengths and 'hear' the reflections, not to use a vision system that passively receives light waves. They shouldn't be affected by ambient light in theory. As of the crash ground -- It seems to be located at the roof of some old machinery room or some kind of facility house with no windows or access, or contact info. It is more than 3 floors tall, so I don't think I can just climb like spiderman. I am running out of my imaginations on how can I access that. --Update: I just went to the crashing site again, and there was a 6-stories parking garage on the other side of the street. I climbed up there and I could see the roof of that crashing site. The drone was not there. Not a single debris or arm or propeller can be seen. Simply nothing on the roof other than dirt and pebbles. The geo location was not accurate, I guess.

Eric L Posted at 12-12 11:44
Any idea why other radar wavelength was not used? And, even visible wavelengths were used, I can't imagine how that could affect the rada

I would guess because
a) the drone does not have forward / backward facing emitters for any frequency of EM and that
b) the OA cameras are receptors of visible light only.

Barring the antenae, the only emitters of EM that I can see are downward looking InfraRed emitters on the bottom of the drone and the also downward looking, auxillary light.

I doubt any drone uses "RADAR", I supect it would need an emitter.

Eric L Posted at 12-12 13:01
Thanks for the analysis, it matched everything we concluded so far. The building under construction is now around 135m/440 feet tall, just a bit higher than RTH 120 m height. Do you know why OA radar can be affected by ambient light? As far as I know, radars emits electromagnetic waves in different wavelengths and 'hear' the reflections, not to use a vision system that passively receives light waves. They shouldn't be affected by ambient light in theory. As of the crash ground -- It seems to be located at the roof of some old machinery room or some kind of facility house with no windows or access, or contact info. It is more than 3 floors tall, so I don't think I can just climb like spiderman. I am running out of my imaginations on how can I access that. Any suggestions or wild ideas?

You have got hold of the wrong information, DJI do not use ANY form of radar for obstacle avoidance, it is a purely visual system. The sensors are actually small cameras.

Sean-bumble-bee Posted at 12-12 13:11
I would guess because
a) the drone does not have forward facing emitters for any frequency of EM and that
b) the OA cameras are receptors of visible light only.

Ah I see, thanks! I assumed drones can equip stuff widely used in vehicles. Plenty of radar technologies have been implemented and being used everyday, but I forgot they are also much more expensive than cheap cameras. They could cost a few hundred dollars themselves so it is not economically wise to implement in Mavic drones. I somehow assumed DJI has found out cheap ways to achieve this.

Eric L Posted at 12-12 12:46
Lessons learnt, in the hard way. I made a series of seemingly trivial mistakes that ultimately led to this incident. I guess it is a lesson that I will have to learn at some point, sooner or later.  I assumed the obstacle avoidance system during RTH was smart enough to not to collide into a building directly in 2021. And obviously that was not a valid assumption, it is still not quite as 'smart' as I imagined from commercials... As the radar was somehow affected by ambient visible wavelength light conditions, something I failed to comprehend even today. As a pilot with plenty of experience flying in open plain but limited experience flying around tall buildings, I have yet to have touched the boundary of how far this obstacle avoidance and RTH can save me. This is a new building under construction, so there is no telling how high it is  before flight.Multiple factors led to this incident and here I am.

I made a series of seemingly trivial mistakes that ultimately led to this incident.
There was nothing at all trivial about:
Flying far out of sight
Flying in the dark
Not maintaining situational awareness
Flying behind a tall building
Not setting an appropriate height to clear obstacles you fly behind
Not being aware of the technology of the drone and assuming it will magically make up for poor flying

Any one of those could lose your drone.
In combination, in your flying environment, the loss was almost guaranteed.

I assumed the obstacle avoidance system during RTH was smart enough to not to collide into a building directly in 2021. And obviously that was not a valid assumption, it is still not quite as 'smart' as I imagined from commercials.
Rather than imagining and assuming, it would be smart to read the manual and find out what technology the drone has and the limitations of that technology.
For your drone it was all explained on p17-18 here:
https://dl.djicdn.com/downloads/ ... anual_v1.0_enIV.pdf

Labroides Posted at 12-12 14:28
I made a series of seemingly trivial mistakes that ultimately led to this incident.
There was nothing at all trivial about:
Flying far out of sight

Well, I don't really know how to respond to your comments.
Nothing from your response was wrong, they are all 100% technically correct. As many pilots above already pointed out, there are plenty of things I did wrong. And I admitted them, I said "I made a series of seemingly trivial mistakes that ultimately led to this incident", not "I made a series of trivial mistakes that ultimately led to this incident."
After the cause was revealed from flight logs, I was trying to explain the incident from my perspective, and how I got it wrong, with a bit of emotions mixed into it. That's all. It's a pretty traumatic experience, and like I said, lessons were learnt. And to be honest with you, the last thing I need here is a smart guy reminding me what an ignorant dumbass pilot I was, however heart-sunking true that is.
Also, if the expectation is every pilot to read a 56 page long user's manual to learn what technologies and each every corner cases before taking off, for a recreational camera drone, then I don't know - As a hobbyist photographer, maybe we should design a shorter version of manual? Or just design the product to have more reasonable safety features so it can be more fool-proof from dumb population like me? Is asking a drone not to collide into a concrete wall at 6pm in a well-lit city downtown too much to ask for DJI for $1299 in 2021? From what I know, Tesla was doing the same visual sensor technology on their self driving package - Is Tesla using better cameras or better algorithms to avoid collisions on highway? Do they tend to collide after sunset? Just curious
That being said, I learnt quite a lot about Air 2S during this incident, and I am sure as hell that I won't repeat the same mistake twice, especially when it's going to cost 1K out of my pocket each time. I appreciate whoever contributed to this investigation. Thank you.

Eric L Posted at 12-12 16:19
Well, I don't really know how to respond to your comments.
Nothing from your response was wrong, they are all 100% technically correct. As many pilots above already pointed out, there are plenty of things I did wrong. And I admitted them, I said "I made a series of seemingly trivial mistakes that ultimately led to this incident", not "I made a series of trivial mistakes that ultimately led to this incident."
After the cause was revealed from flight logs, I was trying to explain the incident from my perspective, and how I got it wrong, with a bit of emotions mixed into it. That's all. It's a pretty traumatic experience, and like I said, lessons were learnt. And to be honest with you, the last thing I need here is a smart guy reminding me what an ignorant dumbass pilot I was, however heart-sunking true that is.

Is asking a drone not to collide into a concrete wall at 6pm in a well-lit city downtown too much to ask for DJI for $1299 in 2021?
Asking the question indicates a poor understanding of drone technology and it's limitations.
It also indicates the idea that technology will or should magically absolve you of the need to fly responsibly and intelligently.
There are 100 things that can go wrong and cause the loss of a drone.
You need to understand those things and how to make sure they don't happen to you.

From what I know, Tesla was doing the same visual sensor technology on their self driving package - Is Tesla using better cameras or better algorithms to avoid collisions on highway?
I don't know or care what Tesla do, but comparison with Tesla is irrelevant.
Their cars cost significantly more than your toy drone and they don't have the same serious weight constraints as the drone manufacturer.

Update: I visited the crash site again, and although the last geo location indicates it fell on to a roof of a 3-story tall local facility, it was actually not there. Nothing can be seen visually, not even a debris. The new construction site next to it is a building much higher than this facility's roof itself, reaching 135m+. There are little distance between the facility building and the new building under construction, so I don't think there is a high chance for drone to land in that slit piece of land either. I don't know where it landed.

Labroides Posted at 12-12 16:29
Is asking a drone not to collide into a concrete wall at 6pm in a well-lit city downtown too much to ask for DJI for $1299 in 2021?
Asking the question indicates a poor understanding of drone technology and it's limitations.
It also indicates the idea that technology will or should magically absolve you of the need to fly responsibly and intelligently.

Asking the question indicates a poor understanding of drone technology and it's limitations.
You already drew this conclusion in your previous comment, didn't you? Thanks for the kind words, but no one needs redudant information repeated over and over again.

I don't know or care what Tesla do, but comparison with Tesla is irrelevant.
Their cars cost significantly more than your toy drone and they don't have the same serious weight constraints as the drone manufacturer.It's quite the opposite. Not only they are relevant, and as a proper consumer, you should care. None of your arguments were valid.

First, the weight. They are built on top of cameras + processor, and as you probably know, they can be afforded on drone (Luckily, we have multiple camera sensors on Air 2S, with extra ultrasonic sensors). As long as we use RCCB layout for the pixels on CMOS, which isn't pricy or heavy, we can achieve decent infrared capabilities to begin with.

Second, the cost. Matter of fact, Tesla used camera based OA to reduce costs, since radars are quite expensive but cameras are dirt cheap these days. As of software 3D modeling based on sensor imaging, I think DJI already have that developed, they just need a CMOS that can sense electromagnetic waves beyond 700 nm. You don't need to manufacture a new software for each device so that shouldn't add too much burden on their developers, so no cost there neither.

Finally, if you are determined to prove I am a poor pilot with little knowledge over the device I own, you have already proved your point, and I admitted it multiple times. So when you reply, try a different angle.

Eric L Posted at 12-12 16:38
Update: I visited the crash site again, and although the last geo location indicates it fell on to a roof of a 3-story tall local facility, it was actually not there. Nothing can be seen visually, not even a debris. The new construction site next to it is a building much higher than this facility's roof itself, reaching 135m+. There are little distance between the facility building and the new building under construction, so I don't think there is a high chance for drone to land in that slit piece of land either. I don't know where it landed.

although the last geo location indicates it fell on to a roof of a 3-story tall local facility
The data doesn't indicate that.
It's not possible to tell from the data, where the drone ended up.

What the data shows is that at 2:34.7 it was at 40.72032  -74.04166 and after that there were no updates to its location data.
First this was because the drone was behind the building and out of contact.
Shortly after, when signal was regained, there was no GPS data because the drone was tumbling and spinning and upside down much of the time.

When the last good location data was received, the drone was 400 ft up and flying at approx 28 mph.
It's unlikely that it fell vertically from the position it was at that time and the place it came to rest is not able to be determined.

If you look at the flight track at the end of the flight on the Phantomhelp page, extend it a bit to hit the hi rise building. If it hit something solid, like a wall, I would suggest that the drone is not too far from a point directly beneath the point of collision. It seems to have suffered a largely uninterrupted fall. If it hit scaffolding and etc and could carry on moving forward then things are more complicated.
The attached show the last few seconds of the flight, it might just have been reaching terminal velocity at the start of the ATTI

Eric L Posted at 12-12 16:38
Update: I visited the crash site again, and although the last geo location indicates it fell on to a roof of a 3-story tall local facility, it was actually not there. Nothing can be seen visually, not even a debris. The new construction site next to it is a building much higher than this facility's roof itself, reaching 135m+. There are little distance between the facility building and the new building under construction, so I don't think there is a high chance for drone to land in that slit piece of land either. I don't know where it landed.

The big thing I noticed from your pictures there from a fire protection person's perspective, the picture on the right towards the left Edge (bottom). Those are large high power Transformers. If your drone would have fallen into those, and caused some kind of electric Arc leading to a fire...oh boy!! Especially if they were somehow to find out that it was your drone that caused it. You would have been on the news. It would have made one hell of a mess of a fire. Definitely should be a wake-up call to never fly anywhere near those again.

I may be opening a can of worms here. But is the OA actually daylight required? I see a lot of IR flash when video recording my drones with an IR device recorder, so why is that not light independent? Just asking.

GaryDoug Posted at 12-12 22:20
I may be opening a can of worms here. But is the OA actually daylight required? I see a lot of IR flash when video recording my drones with an IR device recorder, so why is that not light independent? Just asking.

But is the OA actually daylight required?
There's a clue in his flight data which had multiple warnings that say:
   Ambient light too low. Vision system and obstacle sensing unavailable. Fly with caution

I see a lot of IR flash when video recording my drones with an IR device recorder, so why is that not light independent?
The IR sensors are downward facing and aren't used for obstacle avoidance.

The #1 rule of setting RTH higher than everything in your flight path was not done.  It's a rule that everyone should know but unfortunately isn't taught to everybody.  Having the 400ft law prevent you from not being able to set the RTH correctly made you crash.  If you are scared about breaching 400ft then fly where you can see it and ensure that anything near is not taller than 400ft from your take-off position.  I try to teach everyone this when I show them about drones.  That is the #1 thing to always do besides updating all the firmware, doing all the calibrations, and ensuring you have GPS signal.  I've had jobs where I had to do them without GPS and it is stable but this is because there is little to no wind in the areaI'm flying, I can see it, and I feel confident it responds to All my intents.

I always set proper RTH when in the air (flying as high as the tallest obstacles and setting Higher).  I also ensure to know what the failsafe procedure is when the signal is lost, RTH, hover, land.  I change them depending on what I'm doing.  If I'm flying under trees, you should not use RTH or it will crash into the branches above.  If you are over water, you better not set land unless you are on a boat.  It's so common sense to me but it's 100% understandable when people new to the activity don't know how things should be done.  The logic is good though, not breaching 400ft AGL but ultimately that screwed you.  Who knows if there is an on-going investigation to find the pilot that caused damaged to what it hit.

Some people need a quick education on the limitations of vision OA sensors under low-light.  Some people seem to not run the proper test before they are too late.  Before I do anything with new drones and new activities I run "tests" and I get lucky a lot of times then I try to tell everyone my findings but unfortunately the mediums at which to make "everyone" know doesn't exist so all I can do at times is just read what went wrong.

The Skydio 2 was the greatest tracking drone under non-lowlight and no trees I've ever witnessed.  The live feed is just ultra garbage though.  Check out the insane (literally) tracking at 10:33.  

AntDX316 Posted at 12-13 01:11
The #1 rule of setting RTH higher than everything in your flight path was not done.  It's a rule that everyone should know but unfortunately isn't taught to everybody.  Having the 400ft law prevent you from not being able to set the RTH correctly made you crash.  If you are scared about breaching 400ft then fly where you can see it and ensure that anything near is not taller than 400ft from your take-off position.  I try to teach everyone this when I show them about drones.  That is the #1 thing to always do besides updating all the firmware, doing all the calibrations, and ensuring you have GPS signal.  I've had jobs where I had to do them without GPS and it is stable but this is because there is little to no wind in the areaI'm flying, I can see it, and I feel confident it responds to All my intents.

I always set proper RTH when in the air (flying as high as the tallest obstacles and setting Higher).  I also ensure to know what the failsafe procedure is when the signal is lost, RTH, hover, land.  I change them depending on what I'm doing.  If I'm flying under trees, you should not use RTH or it will crash into the branches above.  If you are over water, you better not set land unless you are on a boat.  It's so common sense to me but it's 100% understandable when people new to the activity don't know how things should be done.  The logic is good though, not breaching 400ft AGL but ultimately that screwed you.  Who knows if there is an on-going investigation to find the pilot that caused damaged to what it hit.

I totally agree. I only hope I've met you guys and be educated before I conducted these flights. I will tell other new pilots and hope they don't repeat my mistakes.

DAYTONA392 Posted at 12-12 20:21
The big thing I noticed from your pictures there from a fire protection person's perspective, the picture on the right towards the left Edge (bottom). Those are large high power Transformers. If your drone would have fallen into those, and caused some kind of electric Arc leading to a fire...oh boy!! Especially if they were somehow to find out that it was your drone that caused it. You would have been on the news. It would have made one hell of a mess of a fire. Definitely should be a wake-up call to never fly anywhere near those again.

Yes. I was complaining about getting hit with a bullet all this time, but I might have as well just dodged a cannon ball. I have been updated on what is "flying responsibly" with a lot of useful tips from lots of senior pilots here, e.g. Set RTH height higher than highest obstacle or never fly behind it, or don't count on OA when it's dark. You're welcome to add more to the list so others can see and hopefully avoid repeating this as well.
As of the transformers' location, let's say... Obviously I won't fly around that particular spot in my lifetime, but do you think that is enough from a fire protection person's perspective? Will other people notice its existence?  Will other inexperienced pilots eventually fly around this spot 10, 20 years down in future? What if they made the same mistake and lost signal? What if some criminal/terrorist deliberately conducts an attack? ... It makes sense to me to set up some fencing on these transformers from top as well, if a drone's body can cause a massive fire. We can't count on every people being intelligent and smart and responsible all the time to be fire safe, that's just not true when it comes to humanity.

Eric L Posted at 12-13 06:35
Yes. I was complaining about getting hit with a bullet all this time, but I might have as well just dodged a cannon ball. I have been updated on what is "flying responsibly" with a lot of useful tips from lots of senior pilots here, e.g. Set RTH height higher than highest obstacle or never fly behind it, or don't count on OA when it's dark. You're welcome to add more to the list so others can see and hopefully avoid repeating this as well.
As of the transformers' location, let's say... Obviously I won't fly around that particular spot in my lifetime, but do you think that is enough from a fire protection person's perspective? Will other people notice its existence?  Will other inexperienced pilots eventually fly around this spot 10, 20 years down in future? What if they made the same mistake and lost signal? What if some criminal/terrorist deliberately conducts an attack? ... It makes sense to me to set up some fencing on these transformers from top as well, if a drone's body can cause a massive fire. We can't count on every people being intelligent and smart and responsible all the time to be fire safe, that's just not true when it comes to humanity.

Not sure you noted that being 1/2 mile away was not within visual line of sight.  That’s a required by law rule.  Lot of your issues  would have been eliminated if you had actually been able to see your drone…,

Eric L Posted at 12-13 06:10
I totally agree. I only hope I've met you guys and be educated before I conducted these flights. I will tell other new pilots and hope they don't repeat my mistakes.

Being able to RTH reliably is the most important reason why I can fly distance with confidence.  With analog and no RTH I don't even leave the park.