P4 Flips Backward On Takeoff

[Ricrhys]

Morning,

after updating the firmware the last two times I have been out with my P4 (standard) its flipped over backwards on take off, not every time though, I was using DJI go4. The first time I put it down to a combination off wind and not the flattest take off spot, admittedly it was a little windy but no more than I would usually fly in. After that first time I recalibrated the IMU then took it out yesterday and it happened for the second time, I held it in the air and took off from there and it flew fine, it's knocked my confidence a bit but I'm wondering if its something I'm doing/not doing?

Thanks in advance

Ric

 

[AeroMirage]

If when using the CSC to launch, you dwell the sticks in and to the center, that is ordering the AC to rotate to the right and move backward.
When I use that to turn on the props, as soon as I see the props begin to rotate, I let go of the sticks and let it idle a second, then move left stich forward to lift off. Or use auto take off.

 

[MGS9500]

You're going to get a fair number of responses, but this issue can be due to a number things.
The good news is I don't believe it is a hardware problem. I noticed that you don't have too many posts, so I assume you might be new to the game. This is more commonly user error.

This problem can be reproduced if:
1.) The ground isn't level.
2.) Failure to confirm that the camera is free of all obstacles.
3.) Failure to apply adequate power when taking off after power on to the propellers.

If you apply power slowly, or don't go into a straight vertical climb on first power, the drone takes off slowly and can't orient itself.

The auto take off will apply perform a quick takeoff to a specific height. When taking off, apply power without any forward/backward toggle. Take off quickly, once the drone reaches 4-10 feet, release the power and the drone will orient and stabilize.

I'm curious if others have added suggestions. Like I said, this isn't typically a hardware issue. New users apply power slowly and this may cause the problem

 

[Ricrhys]

Thanks for the quick replies, I've only used auto take off and I just let it do it's thing till it's hovering and stable, should I be applying power myself during this auto procedure? and yes you guessed correct, this is my first real drone so I'm pretty green!

 

[N017RW]

Why auto-take off?

Just apply 100% throttle till your comfortable with height then release and allow stabilization. If it hovers it should fly fine short of some type of interference warning.

 

[Ricrhys]

Why auto-take off?

Just apply 100% throttle till your comfortable with height then release and allow stabilization. If it hovers it should fly fine short of some type of interference warning.

Not sure really, just kind of assumed auto was the safest way as a new flyer and have done it like that ever since, will give manual a try, to be honest it does seem like in auto it revs up then backs off a little before giving it the beans.

 

[N017RW]

Yea, cool to know it works but I prefer old-school. Guess cause I’ve been an r/c flyer since the 70s.

 

[sar104]

Thanks for the quick replies, I've only used auto take off and I just let it do it's thing till it's hovering and stable, should I be applying power myself during this auto procedure? and yes you guessed correct, this is my first real drone so I'm pretty green!

With auto-takeoff you don't need any other user interaction, so that's not the cause. If it continues to happen I suggest retrieving the .DAT flight log, which should show exactly what is happening.

 

[Ricrhys]

Yeah only it's not working really anymore so I'll give manual a shot tomorrow and see how it goes.

 

[Ricrhys]

With auto-takeoff you don't need any other user interaction, so that's not the cause. If it continues to happen I suggest retrieving the .DAT flight log, which should show exactly what is happening.

Thanks, I'm just downloading them now after googling how, I have no idea what I'm looking for mind in the data.

 

[sar104]

Thanks, I'm just downloading them now after googling how, I have no idea what I'm looking for mind in the data.

Feel free to post them here, or a link if you upload to Dropbox etc. A number of forum members will be able to analyze them for you.

 

[Ricrhys]

Feel free to post them here, or a link if you upload to Dropbox etc. A number of forum members will be able to analyze them for you.

Thanks, the dat file came in at just under 1gb does that seem right? I only downloaded yesterdays flights, airdata web site is saying biggest file size is 15mb.

 

[sar104]

Thanks, the dat file came in at just under 1gb does that seem right? I only downloaded yesterdays flights, airdata web site is saying biggest file size is 15mb.

If that's a collection of DATs with a name of the form DJI_ASSISTANT_EXPORT_FILE.......DAT, then that is quite reasonable.

 

[sar104]

Thanks, the dat file came in at just under 1gb does that seem right? I only downloaded yesterdays flights, airdata web site is saying biggest file size is 15mb.

Note also that AirData doesn't read aircraft DAT files - only the mobile device logs, which are much smaller.

 

[Ricrhys]

Thats the one, I'll get it up somewhere and report back, thank for spending time on this guys.

Four and a half hours to upload on house wifi, forty five minutes using my phone as a hotspot.

 

[Ricrhys]

Here is the link DJI_ASSISTANT_EXPORT_FILE_2018-03-05_15-37-30.DAT I've no idea how to read these files but will do little research now, I've just got in. I'm not 100% sure but I think the flight where it crashed was one of the later ones, I had the drone on and off a lot that morning. Will it be blindingly obvious from reading the files? I'm assuming so as I only had it actually up in the air for the one flight the rest were sat on the dining room table.

 

[sar104]

Here is the link DJI_ASSISTANT_EXPORT_FILE_2018-03-05_15-37-30.DAT I've no idea how to read these files but will do little research now, I've just got in. I'm not 100% sure but I think the flight wherein crashed waste of the later ones, I had the drone on and off a lot that morning. Will it be blindingly obvious from reading the files? I'm assuming so as I only had it actually up in the air for the one flight the rest were sat on the dining room table.

It should be easy to figure out - I'll take a look later today.

 

[sar104]

That's an interesting event. The first takeoff looks like it was simply blown over backwards by the wind:

At around 1.9 s the aircraft starts to tip backwards on lift off, and the FC tries to correct by increasing the rear motor speeds and leveling off the front motor speeds. It does not correct the flip and the aircraft ends up upside down.

The second takeoff doesn't give many clues - it was clearly hand-launched as you mentioned:

 

[Ricrhys]

Thats good news then, it was behaving like it should, thanks again for taking the time to help me out with this.

 

[N017RW]

Full throttle on take offs!