Geebax Posted at 2017-3-1 18:51
Looking at the flight logs, the thing I noticed first was that you began the flight with 52% battery. The last log entry had the battery at 45%. The error message of gimbal roll reaching end limit often occurs under sudden change in direction, and as you were flying in ATTI mode at the time, that sudden movement can also place a heavy load on the battery.

One effect I am convinced occurs, is that the battery percentage indication is not accurate on a partially charged battery, and at several occasions during the flight, the battery voltage was down to 13.8 volts, which is not good.

I tried looking at google earth and you can see a sudden drop after the gimble tilt.

I do not understand what cause this.  Does appear to be over the water but close to a building. Whirlybird winds can certainly come off buildings.  However without a proper GPS fix as not enough satellites picked up it is possible the drone was not in the position on google earth.

My analysis though is mere speculation only based on what I see.  I do not know how DJI make their analysis.  

Geebax Posted at 2017-3-1 18:51
Looking at the flight logs, the thing I noticed first was that you began the flight with 52% battery. The last log entry had the battery at 45%. The error message of gimbal roll reaching end limit often occurs under sudden change in direction, and as you were flying in ATTI mode at the time, that sudden movement can also place a heavy load on the battery.

One effect I am convinced occurs, is that the battery percentage indication is not accurate on a partially charged battery, and at several occasions during the flight, the battery voltage was down to 13.8 volts, which is not good.

What you are saying at least seems logical on paper.

It started this flight at 54%. Is that low... Sure. But I need 2-3 photos of the base of a building. This battery has been used only twice. I'm flying with no winds and over 13minutes on the timer.

Every Indication was the battery was able to support whatever manuver I threw at it.


Again this battery has only been used twice. The aircraft has only seen 7 flights. I have 4 batteries. Battery was fully charged when I took it out. I ran a flight, brought it home at 60% and reviewed video of what I captured, hence the 54%.

Like I said, I'm not a novice to flying drones, I do this around the world for a hobby, and I also do it commercially. I have much confidence in a 50% battery handling these minute manuvers. And I have no problem taking the blame for unneccesary risk or being an overt crash dummy. It has happened before. .

This day was NOT the case.

blackcrusader Posted at 2017-3-1 19:15
I tried looking at google earth and you can see a sudden drop after the gimble tilt.

I do not understand what cause this.  Does appear to be over the water but close to a building. Whirlybird winds can certainly come off buildings.  However without a proper GPS fix as not enough satellites picked up it is possible the drone was not in the position on google earth.

I'm also led to believe that you think it is nearly impossible for a drone to malfunction and fall out the sky.

Does the image below not give you pause? The drone should have collected at least 5+ more seconds of data at the distance I was traveling to the closest obstacle.

Even without Sat data, I had high signal data, which is obviously collected remotely. Data would have been collected. It should not have simply "disappeared". See images below.

AlecW Posted at 2017-3-1 17:51
That made me laugh.  Too cold for birds?

Jokes aside. OP are you not seeing ANY of the pictures I have posted either? I now have at least 15-20 posted. No one has acknowledge seeing them yet.

DJTek Posted at 2017-3-1 19:17
What you are saying at least seems logical on paper.

It started this flight at 54%. Is that low... Sure. But I need 2-3 photos of the base of a building. This battery has been used only twice. I'm flying with no winds and over 13minutes on the timer.

Yes. Since the first posting, all the photos are visible.

If you have been around this forum for a long time, you will have seen dozens of similar scenarios. So many, that the experienced pilots will not hesitate to say to any new flyer: 'Never take off for a flight on a less than fully charged battery'. The fact that your battery was new has nothing to do with the issue.

And at the same time, I have seen many logs that show a sudden high demand on an already depleted battery can caused an aircraft to shut down in flight and fall out of the sky.

I offer you the explanation, but as is your choice, you can take it or leave it.

Geebax Posted at 2017-3-1 19:24
Yes. Since the first posting, all the photos are visible.

If you have been around this forum for a long time, you will have seen dozens of similar scenarios. So many, that the experienced pilots will not hesitate to say to any new flyer: 'Never take off for a flight on a less than fully charged battery'. The fact that your battery was new has nothing to do with the issue.

Thanks for acknowledging the photos.

Honestly you have the most logical explanation. Personally I have to disagree but it's certainly plausable.
Resetting this conversation, the purpose I truly was trying to prove is that it never flew into a building 100feet away as "Assumed" by DJI.

Even with a drone crash, as long as that lightbridge transceiver is still connected, it will send data until it's morbid death. This is different, I was panning left and it tilted and dropped out the sky.

I have faith that DJI's batteries can last 5 minutes on a light load at 50%. Especially knowing when voltage is at around 11v it begins to scream to go home. I appreciate your insight and ability to look at this logically without Bias.

DJTek Posted at 2017-3-1 19:22
Jokes aside. OP are you not seeing ANY of the pictures I have posted either? I now have at least 15-20 posted. No one has acknowledge seeing them yet.

I can see them in the later postings not in the earlier postings.

DJTek Posted at 2017-3-1 19:20
I'm also led to believe that you think it is nearly impossible for a drone to malfunction and fall out the sky.

Does the image below not give you pause? The drone should have collected at least 5+ more seconds of data at the distance I was traveling to the closest obstacle.

Because uo your flying area analysis isn't simple but here's what I can see;

You started the flight in a small relatively open area and flew into a more confined area surrounded by 35 storey buildings.
This blocked a significant portion of the sky and you lost enough satellites to lose GPS position holding at 1:30 and again at 2:00.5 which caused the sat count to drop to 3-4.
When the Phantom does not have enough sats to get a positional fix, it cannot give any position or speed data but it can drift with the wind and will continue on whatever path it was traveling.

In your first illustration (the one using DJI supplied data with your notes in red), we can see that the Phantom was pointing right from your perspective.
At 2:00.5 you lost GPS.
The Phantom at this point is southwards at 6.5 metres/sec and facing westwards.
Your joystick input shows that you held the right joystick to the left from 1:52.4 until the flight record stops at 2:05.4.
At a minimum this would have caused the Phantom to continue on its trajectory for 5 seconds after GPS was lost.

You ask in the red annotation on the first illustration:  Why are you making up data, where is this 100 ft of drifting?

I think the flight record accounts for this 100 feet of drifting (5 seconds at 6.5 m/s = 32.5 metres)

The final message (Gimbal Roll Reached Movement Limit) may be a result of an impact.
The compass indication you saw on HD is almost certainly related to an impact.

DJTek Posted at 2017-3-1 19:20
I'm also led to believe that you think it is nearly impossible for a drone to malfunction and fall out the sky.

Does the image below not give you pause? The drone should have collected at least 5+ more seconds of data at the distance I was traveling to the closest obstacle.

Yes the image gave me pause to think what could have caused that.

I was returning my drone from a flight on RTH and when it was within 30m of the vertical drop next to where I had taken off I put in a yaw command.  My drone did several whirlybird turns, I stopped my yaw command and the drone righted itself.  Was weird as I have done that before with no issues so assume the yaw command with some wind gusts put me into a nice spin.  Sure was fun to watch.  At least I had several hundred feet below to recover my drone.

AlecW Posted at 2017-3-1 19:34
You are an observant guy, sounds fishy.  Oh wait.. I'm on to something.  Googling about fish in the cold, I don't think they swim in Chicago in the winter

You win AlecW, I deduce it's hard to be objective in this group .

For what it is worth.... here is the actual perspective and view that I had when my drone dropped. It's actually not hard to believe that you can hear a splash of a heavy 3 pound object hitting still water.


Oh more importantly as you can see in my video... It's even more challanging to believe that it simply hit a building as the DJI Technical team has assumed (with over 100ft of drifting with no data being sent to the controller)

This is a link for a video - Click on it to watch 30 second annotated  for the DJI Forums Group

DJTek Posted at 2017-3-1 19:48
You win AlecW, I deduce it's hard to be objective in this group .

For what it is worth.... here is the actual perspective and view that I had when my drone dropped. It's actually not hard to believe that you can hear a splash of a heavy 3 pound object hitting still water.

"It's even more challanging to believe that it simply hit a building as the DJI Technical team has assumed (with over 100ft of drifting with no data being sent to the controller)"

I'd suggest checking my post #53 above and see how that matches what happened.
The Phantom had no position or speed data to transmit because it ad lost GPS.
BUT that doesn't mean that no data was being sent.
The Phantom was traveling toward that building when GPS was lost.
The Phantom would not have stopped in place, it would have continued on that trajectory and speed would have been maintained because of the joystick input.

At a distance of at least 700 feet away (perhaps it was 800 ft), I'd be skeptical about your ability to see exactly where it splashed down.

Very sorry to hear about your crash, and we reviewed the flight records since you have query about the analysis result.
Please check the information below:
01:46: GPS lost, the speed was 6.3m/s, the aircraft switched to Atti mode. The coordinate would not refresh since there was no GPS signal, as a result, the yellow line would not show up between the last GPS location and the building where it crashed. Besides, during this period, you kept moving stick to the left.
According to the image, you flied aircraft at night, it would be better to fly aircraft in a more suitable environment to ensure the safety.
01:52: aircraft crashed and lost balance.

Labroides Posted at 2017-3-1 20:21
"It's even more challanging to believe that it simply hit a building as the DJI Technical team has assumed (with over 100ft of drifting with no data being sent to the controller)"

I'd suggest checking my post #53 above and see how that matches what happened.

@Labroides -  I almost want to believe in what you are saying.
We know for a fact that even with data ommitted (such as speed, velocity) data will continue to stream to the controller.

For the sake of argument  let's say it was hurdling towards the building  with a terminal ending of 5 seconds. My last frame was in front of the building, over the river.  I'm using line of sight, can see the UAV lights, and lining up to capture the base of the building.

The drone fell exactly where I saw my last frame. And call me sonic, alien or mutant... There is no sound like the splash of a heavy object hitting the water. I ran to go see if I could recover it somehow... and it was just gone.

Maybe ... just maybe.... it's actually plausible this was mechanical and not operator error, like everyone is attempting to find. It's probable that it didn't crash into a building,.. The battery concept also plausible, but wasn't even in a warning state at the last data capture.

Perhaps, I'm not a new hobbiest but an experienced drone pilot with many video's under his belt to become acclimated to the behavior of DJI suite of drones. We will find out probably tomorrow when DJI reviews my information and get's back to me.

DJTek Posted at 2017-3-1 20:46
@Labroides -  I almost want to believe in what you are saying.
We know for a fact that even with data ommitted (such as speed, velocity) data will continue to stream to the controller.

Not that it helps you any, but I don't believe it crashed into any building either.

DJTek Posted at 2017-3-1 20:46
@Labroides -  I almost want to believe in what you are saying.
We know for a fact that even with data ommitted (such as speed, velocity) data will continue to stream to the controller.

Your Phantom was heading toward that building, it lost GPS and you continued to pilot it in the same direction.
It had to continue and in the time till the end of the flight record it would have covered that distance.
I can see how DJI came up with their explanation - it fits the data.
I can't see any evidence that DJI lied as your thread title suggests and certainly no reason for fellow pilots to push back against anything (except for false claims that DJI lied to you).

The most likely explanation for the incident is that the Phantom collided with the building.
WIth no position holding due to loss of GPS and continued Joystick input, it's hard to imagine any other scenario.
At a distance of 800 feet, it would be very difficult to tell whether or not it clipped the building.

DJI Natalia Posted at 2017-3-1 20:32
Very sorry to hear about your crash, and we reviewed the flight records since you have query about the analysis result.
Please check the information below:
01:46: GPS lost, the speed was 6.3m/s, the aircraft switched to Atti mode. The coordinate would not refresh since there was no GPS signal, as a result, the yellow line would not show up between the last GPS location and the building where it crashed. Besides, during this period, you kept moving stick to the left.

Ahh We spoke today.

I was told your team had an escalation that would review the data.
Listen I am reading what you are saying.... but you are looking at data very interesting here.

01:46: GPS lost, the speed was 6.3m/s, the aircraft switched to Atti mode.
ME: (I AGREE)

DJI: The coordinate would not refresh since there was no GPS signal,
ME: (YES BUT IF THE UNIT WAS JUST CONTINUING IT'S TRAJECTORY IT STILL WOULD HAVE COLLECTED DATA BEFORE THIS "CRASH" AS IT ALREADY WAS)

DJI: as a result, the yellow line would not show up between the last GPS location and the building where it crashed.
ME: (It is in a River... Not on the street below. It didn't ricochet off the side of a building. It SPLASHED into the water)

DJI: Besides, during this period, you kept moving stick to the left.
ME: (I moved to the left because visually I had enough space to get past the building and regain GPS)

DJI: According to the image, you flied aircraft at night, it would be better to fly aircraft in a suitable environment to ensure the safety.
ME: This is an personal opinion. Not a law or rule. This was a commercial project. This was a request to capture the auxiliary of the building at night.

DJI: 01:52 aircraft crashed and lost balance.
ME: Why is it in a river.

Now I have offered to send you the video. You're team has declined. There was also a legal memo sent  which i have purposely omitted from this dialogue which you should have which provides another eyewitness encounter. We are looking at the same exact data. Both systems, and the camera stopped at the exact same time. Timecode data is also in the video.

The drone could have hit the building, lost a prop and flipped and crashed into the river. A prop came of my drone at about 30 and crashed but into a tea hedge. I was filming some tea farms.
It certainly did not crash straight down it came down at an angle.  Yes I understand you would think if it hit the building it may have crashed straight down and not into the water.

blackcrusader Posted at 2017-3-1 21:30
The drone could have hit the building, lost a prop and flipped and crashed into the river. A prop came of my drone at about 30 and crashed but into a tea hedge. I was filming some tea farms.
It certainly did not crash straight down it came down at an angle.  Yes I understand you would think if it hit the building it may have crashed straight down and not into the water.

Bro. Buddy. Big man. Friend.

Listen. I get it. People want to truly believe this is operator error. But I'm connected to my drones. I fly them quite regularly, and make videos damn near every week.

There's nothing more crushing than hearing your drone hit water. I've heard it before. And I've seen my drones crash many times also. I'm connected to my investments. I pay heightened attention when I'm flying in risky environments.

I know cause I saw it with my own eyes what my drone did. It never hit a building. I'm watching it insanely closely. I hear the splash. I rush to the location it was. (2 min away) and it had sunk. If you really drone.... you know your craft.

I know what this wasn't. It was not a building collision.

Cabansail Posted at 2017-3-1 13:49
Could it have collided with something flying? (a bird or bat etc).  At almost 700' away at night it would have been very hard to see what was happening.

I never fly at night but could see how bat or a bird could become a real problem.
.

DJTek Posted at 2017-3-1 21:47
Bro. Buddy. Big man. Friend.

Listen. I get it. People want to truly believe this is operator error. But I'm connected to my drones. I fly them quite regularly, and make videos damn near every week.

I don't assume that is operator error or that perhaps as stated by another a battery failure due to low voltage. Merely surmising what can be possible.  You are standing at the home point I assume when in control of your drone, just on 700ft away and flying your drone above and past a railway track when you lose GPS and go to Atti mode as satellites are lost.  So no RTH possible and your drone command was to fly to the left towards the building before the drone crashed.   DJI have escalated this anyway so we wait to see their reply about the several seconds your drone was flow towards the building under your command.  I certainly question my own ability to tell how close my drone is to a building from 700 feet away at night given you also flew past the railway tracks.  Could it be the metal from the tracks and associated overpass caused the loss of GPS connection after you flew past as well?  

DJI said that your drone was commanded to fly towards the building when telemetry stopped. The drone would have kept going that way under Atti mode if you had lost connection.

I had my drone lose connection on a flight I was doing.  However I was already in RTH mode and a mile from home my App went grey and crashed so shut it down. Also lost RC connection.  But my drone just flew back and I was able to re-establish connection and finish the landing manually.

I have no clue from my flight logs as to why I lost connection. I tried a few times to get it back but could not.  So I just sat and waited and eventually heard my drone when it was a 1000 plus feet above me at home point. I wrote about it in here hoping someone from DJI might be able to advise.  But as drone not lost maybe they will not.


  

blackcrusader Posted at 2017-3-1 22:44
I don't assume that is operator error or that perhaps as stated by another a battery failure due to low voltage. Merely surmising what can be possible.  You are standing at the home point I assume when in control of your drone, just on 700ft away and flying your drone above and past a railway track when you lose GPS and go to Atti mode as satellites are lost.  So no RTH possible and your drone command was to fly to the left towards the building before the drone crashed.   DJI have escalated this anyway so we wait to see their reply about the several seconds your drone was flow towards the building under your command.  I certainly question my own ability to tell how close my drone is to a building from 700 feet away at night given you also flew past the railway tracks.  Could it be the metal from the tracks and associated overpass caused the loss of GPS connection after you flew past as well?  

DJI said that your drone was commanded to fly towards the building when telemetry stopped. The drone would have kept going that way under Atti mode if you had lost connection.

"Could it be the metal from the tracks and associated overpass caused the loss of GPS connection after you flew past as well?  "

No amount of steel will have any effect on GPS unless it's blocking the sky view.
It would only affect the compass if the Phantom was flown very close to the bridge (just a few metres away)

Labroides Posted at 2017-3-1 23:06
"Could it be the metal from the tracks and associated overpass caused the loss of GPS connection after you flew past as well?  "

No amount of steel will have any effect on GPS unless it's blocking the sky view.

Yes I meant compass, regret error thanks for the correction.  Now if the drone lost compass it could easily go into a whirly bird spin. There have been several people who have had this when flying near metal bridges near rivers.  

DJTek Posted at 2017-3-1 20:46
@Labroides -  I almost want to believe in what you are saying.
We know for a fact that even with data ommitted (such as speed, velocity) data will continue to stream to the controller.

"The drone fell exactly where I saw my last frame. And call me sonic, alien or mutant... There is no sound like the splash of a heavy object hitting the water. I ran to go see if I could recover it somehow... and it was just gone. "

If you clipped a building or something attached to a building and broke a propeller then the drone wouldn't fall down vertically, it would splash down some distance from the point of contact.  With no GPS data, the position information can't be trusted.

The only way you are going to prove what happened is to return to the point of your last video frame and take it again, this time collecting the GPS coordinates, if you can get an identical frame using an identical camera without hitting a building then that would be reasonable proof, at least for yourself.

It should be obvious from the rate of turn after contact in the logged data if it was an impact or a more gentle compass or power issue.

@ DJTek:
How did you recover the drone from the river? It must have sunk to the ground...

Eric13 Posted at 2017-3-2 01:51
@ DJTek:
How did you recover the drone from the river? It must have sunk to the ground...

I guess that was a stupid question ;-)
I assumed the drone was recovered and DJI had access to the internal log (.dat) files.
Now I understand that the analysis is based on the .txt files from the app. Sorry...

DJTek Posted at 2017-3-1 21:16
Ahh We spoke today.

I was told your team had an escalation that would review the data.

Look man this is very tough to take, I can only imagine, it seems dji and labroids are almost on the same page here.

You had flown a previous flight, so you would or should have been aware of some lack of signal, this is your responsibility.
You took off on a much depleted battery, and at the same time claim to be an experience and also commercial pilot, you will know from your training that this is a no no, your responsibility.
As a commercial pilot you should have had a plan for flying in such a precarious location, you don't mention this, it's your responsibility.
Being a commercial pilot you should have done a safety scenario you don't mention this, your responsibility.
When your Aircraft crashed you had no VLOS as you said earlier you didn't see it crash but heard it. It's your responsibility to have VLOS.
Lastly your a commercial pilot so I have to presume you have the proper insurance in place for such an undertaking, it's your responsibility.

Sometimes we have to take it on the chin when stuff goes wrong, and put it down to experience. But we must always accept responsibility if we are the person in charge.
I hope your insurance coughs up and your back flying soon. Good luck..

If it hit a building why is there no warning saying prop obstruction in the data, there would have been time while it was falling to send that information back to the controller. Even if the impact caused the battery to separate it still should have reported prop obstruction in the data.  
Also why is everyone on this forum so eager to blame the pilot. No one ever reads posts properly, if your going to take the time to respond then at least take the time to read all the facts instead of the half baked bullsh!t responses saying you didn't do this when the OP clearly says they did. FFS.

mic75 Posted at 2017-3-2 04:26
If it hit a building why is there no warning saying prop obstruction in the data, there would have been time while it was falling to send that information back to the controller. Even if the impact caused the battery to separate it still should have reported prop obstruction in the data.  
Also why is everyone on this forum so eager to blame the pilot. No one ever reads posts properly, if your going to take the time to respond then at least take the time to read all the facts instead of the half baked bullsh!t responses saying you didn't do this when the OP clearly says they did. FFS.

If it hit anything including water, im sure you will concede prop obstruction would have been in the data, are you saying it's still flying around.

hallmark007 Posted at 2017-3-2 04:53
If it hit anything including water, im sure you will concede prop obstruction would have been in the data, are you saying it's still flying around.

No I'm saying if it hit a building before falling into the water there would be prop obstruction then a drop in altitude. Still flying around as if.

mic75 Posted at 2017-3-2 04:59
No I'm saying if it hit a building before falling into the water there would be prop obstruction then a drop in altitude. Still flying around as if.

I think the problem there that HD logs may not show this, I would think this last minute data will only be in the dji data file.

I really don't think everyone is blaming OP for this without reason, and many have read very careful circumstances of the crash including dji.

I admit it's very hard to prove malfunction and the weight of evidence laid out here doesn't easily show malfunction.

hallmark007 Posted at 2017-3-2 05:10
I think the problem there that HD logs may not show this, I would think this last minute data will only be in the dji data file.

I really don't think everyone is blaming OP for this without reason, and many have read very careful circumstances of the crash including dji.

Prop obstruction will show up in flight records in the app, while some do read all some don't n there's really no excuse for it. Also if u look at the video of the location, when he was facing the building n made a right stick to the left input there's nothing to his left, only the buildings behind.

blackcrusader Posted at 2017-3-1 18:08
Thanks for uploading the correct original file. http://www.phantomhelp.com/LogViewer/6HJX3ARJ1TCN10BANP01/

Several warnings. Satellite Positioning Off. Fly with caution.  Flying in ATTI mode the drone can easily drift.

Looking at the pitch,roll,yaw figures at the end of your log it certainly looks like a collision into something solid.

A case for insurance rather than warrantee.

DJI states:  "Always launch with a fully charged battery."

Pilot decides to launch his Phantom with 52% battery and promptly flies over water.

Phantom goes down into the water.

Pilot is upset and thinks DJI should pay.

How many times have we seen this?  I want to say hundreds but it's probably only been seven or eight dozen.  

This bird went down due to pilot error.  That's all there is to it.  I know that sounds harsh but it's a fact.  

Sorry for your loss.  

/thread

mic75 Posted at 2017-3-2 05:28
Prop obstruction will show up in flight records in the app, while some do read all some don't n there's really no excuse for it. Also if u look at the video of the location, when he was facing the building n made a right stick to the left input there's nothing to his left, only the buildings behind.

I think there can be many different views your right about that.

But in your initial post you were saying people were spouting BS, but the truth is the OP's first post was vert aggressive accusing dji of lying and being liars, there was a certain amount of BS in that opening post, and it's hard to defend that.
I think he would have had a more sympathetic support if he approached it a bit better. Somewhere along the line we all have to accept responsibility for our own actions, I don't think is doing this, so it's easy for some to get on his back.
I do hope something works out for him, maybe he has insurance or he can do a deal with dji, but at the moment for dji it would be like doing a deal with the devil..

Sorry for your loss!
I have read all the posts in this thread and it seems that there are strong points from "both" ends but ...
If DJI has decided to not replace the craft, then they will not. They are based on their findings and for sure they can find more reasons not to.
Just a suggestion if i may. Maybe DJI can provide you with a discount coupon, you buy a new one and live happily ever after ... hopefully with no major problems!

hallmark007 Posted at 2017-3-2 05:53
I think there can be many different views your right about that.

But in your initial post you were saying people were spouting BS, but the truth is the OP's first post was vert aggressive accusing dji of lying and being liars, there was a certain amount of BS in that opening post, and it's hard to defend that.

This exactly. He comes in here with strong accusations, and then he can't even back them up. There are holes all through his analysis. He is flying at night which is risky, yet he takes no responsibility, he is flying over water with a low battery, yet he takes no responsibility. He thinks he can see everything that is happening from 2 football fields + 100 more feet away, when it's dark out, I call bs. He loses GPS because of his choice of location, yet he takes no responsibility.

If this was me flying and this happened, I would have started a thread where I admitted that this crash was most likely my fault. I would have owned up to the risky aspects of this flight, and I would have simply asked for,help to determine what happened.  

Are you sure there are no obstructions in the area (cables, poles,  etc)?