2nd flyaway and crash in 8 months - BEWARE and install SCR!!

[dalebb]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

good analysis... IMU failure sounds like the most likely cause of the crash.

DJI needs to helps you out getting a new one, they should since it is under warranty.

Update form my side:
I was asked by seller to fill in RMA form and send phantom to DJI in Germany for investigation. I'm not sure I would like it replaced and I would prefer to get full refund and wait for some time, possibly purchasing V3 latter (as we know they solved/improved much more areas than they are officially advertising) or maybe looking for any alternative. I was checking warranty terms and it's not stated what is the period for the whole Phantom (e.g. it crashes), but only separate parts, which looks strange, isn't it? http://www.dji.com/service/warranty-period

BTW, when I purchased first 2 Phantoms which crashed I was looking for the cheapest option in the market (Europe). Can you share your experience with Amazon as a seller, because my understanding is that they treat customers much more better in general and maybe have even better support/response from DJI as Amazon is a big player. I think I saw somewhere that Amazon even accepts the loss to their credit and refunds easier. My question: is it worth to purchase from Amazon where the price is 100-200 USD higher, but this would possibly give much better support/feedback in case of troubles/warranty?

I had this problem too, but mine recovered on auto flight (ground station). I sent mine back to Amazon for full refund.
http://youtu.be/DR5ktSzkEzg

 

[flyNfrank]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

Waiting until a little later...We will probably be involved with a different DJI model. And of coarse we'll be going through all the head aches trying figure out why this does that, ect.

We have had very recent rumors of a new unit due to come out. And when you see a company having a sale on a product week after week it begins to add up.

Edit: I fixed my missed info.

 

[max]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

I had this problem too, but mine recovered on auto flight (ground station). I sent mine back to Amazon for full refund.
http://youtu.be/DR5ktSzkEzg

Wow, crazy video. Does look very similar indeed.

 

[dalebb]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

I had this problem too, but mine recovered on auto flight (ground station). I sent mine back to Amazon for full refund.
http://youtu.be/DR5ktSzkEzg

Wow, crazy video. Does look very similar indeed.

It does, when I put it on the kitchen table and idled it up I could hear (I believe it was the right rear motor) not staying insync with the rest of the motors.

 

[yorlik]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

BEWARE: If you use SCR free, it MAY limit recording length to 3 minutes max! I am still investigating, but this is first time I actuallyl used it to record a flight: it only recorded 3 minutes of my 8 minute flight then shut off. I am sure my screen did not go off at this time (auto shut off by screen blanking)....

 

[yorlik]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

confirmed. 3 min max unless you buy it. In addition, it puts its little icon right over the number of satellites display so time to try another one....

Anyone else have a favorite free screen recorder?

 

[yorlik]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

I need to post a correction to my comments on what the 'radar' pix shows.... I assumed they made it like a regular aircraft and the artificial horizon line showed pitch and roll but no.... from finally recording myself flying with SCR screen recorder, I can tell you they have made it mean something different... This changes the analysis of the OP screen shots some....

I admitted to being dumb in another post, so take this for what it is worth. I am learning and I know I have a LOT to learn still. I also do not lie, like some other member on here who said I called him dumb which I did not.... I think it is a real shame someone feels the need to lie to boost their ego.... I also now know about K-index, which can effect satellite lock on - maybe my case, even tho that other member decided I was REALLY dumb stating I have variables in my satellite lock, based on something.... anyway, sorry, back to the story....

DJI shows roll the same as expected: as the artificial horizon tilts left or right from pure horizontal, it appears to mean that. Ie., if you push right stick full left, it will show artificial horizon tilt to the left the same amount. Ditto right.

But they seem to deviate from std artificial horizon in up vs down. On std aircraft I believe this show rate of climb or fall: aka Pitch.

On DJI it definitely does NOT show that; instead, it shows SPEED forward or backwards! So if artificial horizon line goes above center, it shows speed going straight ahead forward. If it goes below, it is not telling us the drone is falling, but instead, it is moving backwards...

Sure wish they would explain the details of this radar unit so we don't have to figure it out empirically in the field....

 

[max]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

I need to post a correction to my comments on what the 'radar' pix shows.... I assumed they made it like a regular aircraft and the artificial horizon line showed pitch and roll but no.... from finally recording myself flying with SCR screen recorder, I can tell you they have made it mean something different... This changes the analysis of the OP screen shots some....

On DJI it definitely does NOT show that; instead, it shows SPEED forward or backwards! So if artificial horizon line goes above center, it shows speed going straight ahead forward. If it goes below, it is not telling us the drone is falling, but instead, it is moving backwards...

Don't forget that, unlike an airplane, the speed of the quadcopter in the forward or back directions is directly coupled to its tilt. It needs to tilt forward to go forward and visa versa. I haven't looked at the app graphic closely, but I would wager that it does indeed reflect the angle/tilt of the quad. It might be deceiving because the gimbal steadies the camera such that you can't see the effect of the tilt in the video.

Again, I'm not sure, but I would guess that this is what is happening.

 

[artis]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

Hi, I'm really sorry that your bird fell from the sky like that...

I was assembling DIY drones for a year now and I have seen plenty of drone failures.

I have watched your video and from what I see I think it was not a GPS failure nor it was a failed motor/esc. I will explain what in my opinion would have happened in these cases and what I think the real cause is.

[*]GPS problem. In this case the drone would indeed try to fly away to a point it thinks it have to fly, but in a very different manner. The max speed cap is nowhere near 30 meters a second this is simply an impossible speed for a working drone with messed up GPS.
[*]ESC/Motor failure. This option is also highly unlikely. every drone needs at least 4 motors spinning in order to not spin on any of its 3 axis (x, y and z). A failed motor/propeller/esc would send it in a while spin rather than such a high speed dive.

As you can see on the video the gimbal was tilted beyond the drone's max designed tilt/roll angle

The video really looks like to a problem I had in the past with DIY drone.

It was a sunny calm day half a year ago. I went out for a normal fpv flight at about 80 meters. Suddenly the drone started drifting very quickly to the right with no control input to provoke it. I immediately switched off GPS mode , but the problem remained. I pressed my right stick all the way to the left and managed to level the drone but it was still drifting because of its high inertia at that point. I quickly figured that the drone was not able to roll left and able to turn right much more than originally designed. What I did was turn it 180 degrees and use that violent eight roll to bring it back. It worked, but as you can imagine it's hard to land a drone with only one side roll control from 80 meters. I did something like a spiral to bring it down. Suddenly at 40 meters the "mysterious" force that rolled my drone to the right disappeared. I quickly realised that this is my chance to land it quickly. And that's what I started doing. Unfortunately ot about 15 or 20 meters the problem rolled my drone to the right again and since I was in a high speed descent I simply didn't have enough altitude to perform the same maneuver and it crashed on the ground.

Fortunately the drone lost the propellers and slightly damaged 1 motor.

When I hooked up the drone to my PC I was clueless for hours.

Finally I desided to try and recreate some flying conditions on the ground while hooked up to the PC with propellers off and motors spinning at 50 percent. I started turning and rolling it as it would do in a normal flight. I even tried to recreate slight wind blasts which could struck the drone at high altitude. And at this point the killer showed it's face. The horizon went 45 degrees to the left. Since in attitude mode the drone is designed to level itself and since it's horizon was rolled 4t degrees to the left it rolls 45 degrees to the right, not realising it's true horizon level.

This sharp roll angle will cause in very high speed gain followed by the inevitable altitude drop because the drone cannot maintain its lift in this angle and speed.

In my case it was caused by a failed accelerometer (the sensor that is responsible for the horizon detection).

In this case you are screwed. There is almost nothing you could do to recover the drone unless your drone is preconfigured with manual mode on S1 switched and you have one hell of reflexes and considerable experience in flying in manual mode. And even then it is not sure you would have saved it because some platforms (I think naza included) partially use the accelerometer even during manual flight which will level the drone when you release the sticks but will allow you to roll beyond safe limits even allowing you to do a flip.

In your configuration of S1 switch (GPS ATTI ATTI) you simply couldn't do anything to prevent it. So if you feel sorry about doing or not doing something - don't be. There really wasn't anything you could do to save it.

P.S. This could happen because of 2 things. Non calibrated Accelerometer or Accelerometer malfunction. This usually happens in event of a crash but could be manufacturing problem. If you have crashed your bird in the past even slightly, it might be the cause. Accelerometer and Gyro are really sensitive to impacts.

I hope this info helps and I hope DJI will compensate you in an acceptable way.

Good luck.

 

[yorlik]

Re: 2nd flyaway and crash in 8 months - BEWARE and install S

Don't forget that, unlike an airplane, the speed of the quadcopter in the forward or back directions is directly coupled to its tilt. It needs to tilt forward to go forward and visa versa. I haven't looked at the app graphic closely, but I would wager that it does indeed reflect the angle/tilt of the quad. It might be deceiving because the gimbal steadies the camera such that you can't see the effect of the tilt in the video.

Again, I'm not sure, but I would guess that this is what is happening.

Thank you Max for that correction. Since upgrading to 4mm $ 4 gps antenna, I can now get full gps lock in my inside office so can experiment more. I recorded my screen showing radar pitch & yes tilt as you explained. I tried to record it in PFV mode (settings) but that feature does not seem to work - even picked, the camera tries to stabilize my motion. But it is enough to see the range and calibration of the pitch and tilt radar display. Interesting to note that pitch is real angle and tilt (up/down) is log type scale where almost all the way up and down is 45 degree tilt while full 180 is top/bottom.

I thought this might be interesting for others who may not have realized the IMU feedback info is being displayed real time for us in the radar display:

https://app.box.com/s/5mk2ix1ao1c4zf1igmn5ib9ybt90pqkp

As another idea, I wondered what it would show in a full 360 degree tilt so recorded that; as expected, the radar display does go upside down and full around:

https://app.box.com/s/diynx7wmm6dxewwxrzyg287pdxg10rp6

 

[LarasLTU]

Finally I have the update about my case. It took 3 weeks from the moment they received it (total 6 weeks without drone), it was repaired in Germany DJI, which serves Europe. They didn't have any questions and fixed it immediately after it was assigned to technician, according to the defect document they replaced the craft itself, battery - both used, camera, gimbal - not sure if those are new...., also new props. It took 0,75 hours in total according to the document, it also includes the testing I made short test and everything worked fine, I could get even 640x480 resolution without issues - i wasn't able to get it with the old one - I assume Wifi module is still operating fine on the replaced one. In general I can't say any bad word about DJI service in Germany, unless long waiting time until it's assigned to technician, but I was informed that it takes 15 to 20 days as they are serving whole Europe from there. I was hoping, that I will get it replaced and maybe will have an option to get V3, however the lady mentioned, that they replace it 1 to 1 and they can't make other decision unless they get direction from DJI China. I'm not sure if the outcome would be the same if I wouldn't have my phone screed recorded, so I want to highlight this again: use Screen Recorder to capture the sequence of your activities, parameters and video feed from phantom camera.