Phantom 3 malfunction and crash (flight record video)

[blade strike]

From 100 to 400 feet, on every flight I've ever done, I've used 100% up and down for video purposes. Once you get closer to the ground, you need to be aware of the wind it creates, etc. - but where on earth are you guys hearing that you can't move the sticks up or down at 100%? The quad is already "gimped" to do that movement at a slow rate of speed... this is an extremely stable quad literally designed to make these movements.

Other than the opening statue of liberty shot, every single shot in this real required a 100% stick movement one way, while panning the camera at a slower rate in the opposite direction. This isn't user-error, it's how you use a quad to shoot video.

You should really read up on aerodynamics after that kind of commnet.. Vortex ring is real and you should never drop straight down into your dirty prop wash for extended periods of time.

The internal logs might shed some more light onto this one. BTW can you post onboard footage, you did say you were working on jello issues?

 

[leekiya]

Going up fast is ok, but coming down full beans is not recommend in my opinion.

 

[Ttito9]

Listen to some people it's a lost like a family member I'm sorry for your lost I mean if it's something you did wrong while in flight then that's on you but if one of the motors stop working yea I think DJI needs to hook you up with a new one

 

[Adam Kontras]

You should really read up on aerodynamics after that kind of commnet.. Vortex ring is real and you should never drop straight down into your dirty prop wash for extended periods of time.

The internal logs might shed some more light onto this one. BTW can you post onboard footage, you did say you were working on jello issues?

Guys, this thing descends at half the speed of most remote controlled copters. On purpose. But you all seem to be missing what's happening in the flight log:

A) I descend from 144 ft to 90 ft in a little under 5 seconds at 100% - copter is fine.
B) Spin the copter to face me for about 1 second, then correct it at 35% - copter is fine.
C) I began to bring it towards me at around 87% - copter is fine
D) It isn't until I'm doing 2 stick movements at the same time that something goes wrong. I just start to bring it down while going forward and it spins. I let go of the right stick first and the left stick almost immediately after. Something happened with one of the motors, or maybe some balance issue I had with the jello simply got worse? Like the jello was a sign of the malfunctioning unit that progressively got worse? Because during the day? Jello was getting almost unbearable. Which is why I was testing it. It was as if one of the props was "off" and was causing a shake during certain movements... but it never occurred to me that that imbalance could bring down the entire unit. That's crazy... and something everyone with "jello" should be aware of. I did all the calibrations, I was in the process of testing it before a shoot later next week because it renders the unit unusable.

 

[blade strike]

Guys, this thing descends at half the speed of most remote controlled copters. On purpose. But you all seem to be missing what's happening in the flight log:

A) I descend from 144 ft to 90 ft in a little under 5 seconds at 100% - copter is fine.
B) Spin the copter to face me for about 1 second, then correct it at 35% - copter is fine.
C) I began to bring it towards me at around 87% - copter is fine
D) It isn't until I'm doing 2 stick movements at the same time that something goes wrong. I just start to bring it down while going forward and it spins. I let go of the right stick first and the left stick almost immediately after. Something happened with one of the motors, or maybe some balance issue I had with the jello simply got worse? Like the jello was a sign of the malfunctioning unit that progressively got worse? Because during the day? Jello was getting almost unbearable. Which is why I was testing it. It was as if one of the props was "off" and was causing a shake during certain movements... but it never occurred to me that that imbalance could bring down the entire unit. That's crazy... and something everyone with "jello" should be aware of. I did all the calibrations, I was in the process of testing it before a shoot later next week because it renders the unit unusable.

And vortex state can still be an issue. Not saying this was or is the issue. Just trying to educate you a little... This can happen in single rotor craft as well.

Just dont touch the unit and call support first thing on Tuesday if your in the US.

 

[Adam Kontras]

Alright, here's an example of the jello I was dealing with on the last flight...

I think this actually proves something was off before it went out of control. I was never getting this type of jello while hovering, it was only during complex movements...

...then it was during ANY movement...

...and as you can see the day it crashed? I was getting jello while HOVERING. That's got to be a sign that one of those motors was malfunctioning and getting worse. Which also means that others that are dealing with this imbalance issue that causes jello? They need to get that worked out before putting it back in the air.

 

[discv]

Ok.. This would not be motor or esc failure, quads instantly flip.

I would not try to pull the internal logs. I wouldn't touch a thing. Get it into DJI so they can diagnose what happened.

If you could edit the title that would be great? This by no means was a fly away..

That's exactly what the original post seemed to indicate- it did instantly flip. Many times on the way down.
With respect, what do you imagine would happen after the initial 'instant flip'?

 

[Adam Kontras]

That's exactly what the original post seemed to indicate- it did instantly flip. Many times on the way down.
With respect, what do you imagine would happen after the initial 'instant flip'?

IT NEVER FLIPPED!

Let me repeat that! It NEVER flipped. It started to spin (honestly, like someone was behind me controlling it) and it was rocking violently side to side trying to right itself. Absolutely haywire. Imagine every possible stick movement happening at once with 5 controllers. That's what it looked like. But it never flipped over. Had it done that, it would've dropped STRAIGHT down immediately. This struggled for quite awhile.

 

[blade strike]

That's exactly what the original post seemed to indicate- it did instantly flip. Many times on the way down.
With respect, what do you imagine would happen after the initial 'instant flip'?

I guess you read a different post than I did...?

 

[Bebeone]

I've seen on the video of the wreck, that a part of one prop is missing. Did it happend on impact, have you found the missing piece?

 

[mad in nc]

sorry this happened to ya Adam... I really liked the Vegas Videos you posted taken with the bird

 

[Dmacken1]

Alright, here's an example of the jello I was dealing with on the last flight...

I think this actually proves something was off before it went out of control. I was never getting this type of jello while hovering, it was only during complex movements...

...then it was during ANY movement...

...and as you can see the day it crashed? I was getting jello while HOVERING. That's got to be a sign that one of those motors was malfunctioning and getting worse. Which also means that others that are dealing with this imbalance issue that causes jello? They need to get that worked out before putting it back in the air.

I would think that jello is caused by the camera, gimbal or recording and not related to what is going on during a normal controlled flight with average weather conditions.

 

[windexh8er]

The good thing is that this was bought from Advexure, the bad thing is it's still going to be bound by DJI support at some level. My P3P was broke in box (just the controller antenna - cracked) and it's been now roughly 2 weeks since DJI has had my returned P3P and no refund as of yet. Roughly 30 emails have been sent back and forth and I did get someone from the CA office on the phone last week, whom was generally nice, but told me I didn't follow the right process even though I had proven to him the process I followed was the one the first person on the phone told me to follow (which was also bolstered by the email trail of their requests).

The short of it is: DJI support sucks. Be glad you have Advexure as your buffer because they genuinely seem to care in my brief interactions with them. Regardless please track interactions with Advexure and DJI and if you could provide any feedback on timelines that'd be awesome. This is where I think 3DR is going to kick DJI in the nuts - phenomenal support, and that is worth a lot. That Forbes article seemed to indicate Wang (DJI) is hiring more support folks, but at the end of the day it doesn't matter if you hire 100 crappy support people following the same crappy process.

Good luck in the process of replacement, this seems (to me) a clear case of a hardware/software malfunction. But that's just my arm chair assessment, I'll sit back and watch with some popcorn now and keep checking my bank account for the $1300 DJI owes me...

 

[IflyinWY]

Hey Adam, sorry for your loss and, as many have said before me, I hope this turns out in your favor.

For your benefit, and that of the posters who have not read every single post here, perhaps you could summarize your experience as we approach page 8 of this thread.
I too would like to know if you have recovered the missing portion of the prop. I would also like to know what you have surmised as a result of your witnessing, plus your visual inspection of the Phantom since the crash.

Thank you for taking the time to work with us on this issue.

I hope you get another machine and continue to make awesome videos. **** nice work Adam.

 

[discv]

IT NEVER FLIPPED!

Let me repeat that! It NEVER flipped. It started to spin (honestly, like someone was behind me controlling it) and it was rocking violently side to side trying to right itself. Absolutely haywire. Imagine every possible stick movement happening at once with 5 controllers. That's what it looked like. But it never flipped over. Had it done that, it would've dropped STRAIGHT down immediately. This struggled for quite awhile.

Hands up! I misunderstood and got that wrong- and I retract. I also concede that this was not motor/ESC failure in the normal sense.

 

[Oso]

Alright lots of questions, I'll try to answer them all in one post:

-I'm fairly certain you guys are right about one corner motor suddenly either going dead or malfunctioning. It's just crazy because it wasn't during any whacky maneuvering. You can see from the video it sat idle in the sky for a period of time, I turned it and it wasn't going but 2-3 mph when it went haywire.

-yes, the craft was spinning, but not in a perfect circle like it shows in the flight record of course... it was rocking left and right. I mean, just imagine all hell breaking loose when a real helicpoter goes down. That's what it looked like. I wasn't attempting to right it either... it was VERY clear this was over I was watching just to see where it would end up.

-looking at the flight record it was going in and out of atti mode, but no i didn't notice that during, only looking at the video later.

-I couldn't find an angle to take a good picture (cause there was so much damage) so I just took another video:

-I'm gonna attempt to get the .txt files as was suggested now...

Thanks for sharing the video. As I was watching, I was thinking to myself that this must have been a hard video for you to shoot. Then you said it was the worst video you had ever taken. I believe that it was.

I hope you can see beyond the occasional troll to see that most of us feel badly for you and hope that DJI sends you a replacement in exchange. I'm sure they could gain valuable info from studying your crashed bird and its data files.

 

[Dmacken1]

Just a thought here that came to me thinking about this post. I wonder if it is possible to accidentally stop the mechanical mode switch in between two settings at a point where it might ...flutter between the 2 settings.
Again just a thought that came to mind most likely unrelated to this unfortunate crash.

 

[leekiya]

I'm thinking maybe this has got something to do with the craft loosing GPS lock and going into P ATTI mode, all though from what i can see there were no stick rotation inputs made to make the craft start spinning after or just before it lost GPS. There are a few reports coming in from others tho about loosing GPS lock, and it has happened to me. I had 16 locked in and then bam gone. Could be one school of thought to the crash maybe.

 

[aburkefl]

This looks like pro, motor, or ESC failure to me too. If you upload the flight logs (.txt) files from your tablet/phone, I can decode them and export all the data to a CSV file. I can also to some extent decode the FLY*.DAT files located on the internal SD card, use the Pilot app to put your P3 in "flight data recorder" mode then connect to a computer by USB to retrieve them.

I've been able to access these .dat files as well, but apparently no one can read them except DJI and they're not talking about it. There are some flight data files (they're listed as .txt files but they are NOT text data) on my mobile device. They can be converted to CSV files and opened with Excel. The conversion to .csv takes place via www.flylitchi.com.

 

[aburkefl]

I've entered flight data mode a couple times now, it says it's ready to connect but when I plug it in, my computer doesn't see it at all. Maybe something was damaged?

When you enter Flight Data Mode, you need to connect the appropriate cable to the front USB port on the Phantom. That's the USB port covered up by a small, rubberized tab. It's at the front of the Phantom, underneath the label. Unfortunately, as I posted elsewhere here, you can retrieve the .dat files, but they are in a non-readable format. They do NOT work with FLYLITCHY to convert to .csv format. If they can still be read by DJI, they're pretty mum about how.