Bummed. New P3A Crashed. Pls Advise

[TheRealNick]

I'm beta testing our next firmware release and next week should be app upgrade which ties up the time I spend here on pp.com. This thread caught my attention due to all the replies. Before entering I thought, I bet there is some disagreeing and arguing going on, or some naked women pic's in this thread. I see I got half of it right.

I couldn't understand why it took so long for someone to help the OP attach the .txt file. There was so much speculation up to that point.

So I did spend a little time researching the OP's Flight Log. I was able to come up with a Reason for the his crash, but there is no way to know the actual cause due to the reason. The reason for the crash is the R/C. While studying the flight itself there were several points in which the A/C was traveling at consistent speeds. However, there was NO data input from the Sticks. Prior to coming across the R/C situation I was looking at if it was a loss of prop and there just wasn't enough data to support that event taking place.

It is very important to do all calibrations at the time any firmware is upgraded. The IMU reads the X,Y,Z speed, and the data from the compass and GPS. Making sure the things are properly calibrated are the pilot's number one priority. Don't make the mistake and do what someone else is doing. If someone say they have not calibrated the compass in 4 months, so what. That is that guy and his aircraft. Do what is best for you and your aircraft where you live. Here is some good ready on the IMU and Compass: DJI Forum|Explain Compass and IMU purpose ...

I don't understand what you mean as the cause R/C, what is that? What is A/C? ....the roll and pitch abruptly change at the same point with no stick input....how does that not support a loss of 1 motor or prop?

Is R/C the radio controller and A/C the aircraft?

 

[ianwood]

This looks very much like a physical loss of lift / stability i.e. something caused one or more propellers to stop generating sufficient lift. Anything more specific than that is more speculation than not.

To confirm this, there is additional data that we need. We have pitch and roll but I don't see commanded pitch and roll or stick pitch and roll. The comparison of actual, commanded, and stick is very important. If we see the FC attempting to correct external pitch and roll, then it's clear it is a loss of controlled lift event.

 

[DAP-UAV]

Could someone please tell me exactly how to access the flight logs other than the one included in the DJI GO app with the overplayed map. Where is the raw data? How do I download the data? Do I need another app? Please be specific.

Thanks!

 

[flyNfrank]

Could someone please tell me exactly how to access the flight logs other than the one included in the DJI GO app with the overplayed map. Where is the raw data? How do I download the data? Do I need another app? Please be specific.

Thanks!

If you read the manual you will get some idea how to get to what you call the raw data. To use that data, there just isn't much available yet to use the formulas.

Most have been relying on the .TXT file you can download from your mobile device.

 

[flyNfrank]

This looks very much like a physical loss of lift / stability i.e. something caused one or more propellers to stop generating sufficient lift. Anything more specific than that is more speculation than not.

To confirm this, there is additional data that we need. We have pitch and roll but I don't see commanded pitch and roll or stick pitch and roll. The comparison of actual, commanded, and stick is very important. If we see the FC attempting to correct external pitch and roll, then it's clear it is a loss of controlled lift event.

Use the .txt file attached a few pages back.

My speculation is based on the pitch characteristics during the short flight is with the cause being from csc. Pay attention to the height formulas at each tenth of a second can help show the A/C was going down powerless.

 

[McVenturesRC]

Finger tight IS tight enough. but "finger tight" truely MEANS nothing. I guarantee my finger tight in nothing close to what yours is! Or anyone else for that matter! Its a far too subjective term that is a pretty crucial measurement.

So "finger tight" as I would mean is "spin" til it stops, hold motor, lightly snug tighten the prop. Done.

That's what I did.

 

[flyNfrank]

I snug my down with the wrench. Not snug but, Snug to Tight overall. If you go too tight they can strip threads.

 

[TheRealNick]

Use the .txt file attached a few pages back.

My speculation is based on the pitch characteristics during the short flight is with the cause being from csc. Pay attention to the height formulas at each tenth of a second can help show the A/C was going down powerless.

The current was high all the way to impact, could not be CSC....

 

[flyNfrank]

The current was high all the way to impact, could not be CSC....

That's correct. I forgot I looked at that.

Edit: Btw McVenturesRC, were you flying backwards near the time it dropped?

 

[TheRealNick]

In the above picture I paused the playback right at the start of the decent. You can see in the middle picture the phantom banked hard right. In the left picture you can see the phantom nose went down. This matches loss of lift on the front right corner, which matches the corner that had the only intact prop found detached.

From the video in post 140 it is pretty hard to make the props come off and McVentures said he did finger tight. I think if the torque was that critical we would see this all the time. I think something caused that motor to stop. Maybe not motor failure, a logic or esc fault perhaps, who knows...

 

[flyNfrank]

I can barely to not really see the data amounts shown under each pic. The image is now quite large enough. Also before mentioning anything else, where did you pull the data from you are using to reference with? Which file?

 

[TheRealNick]

It is from the video posted in post 383. Looks like he made it from the .csv file that was posted. mcventures also sent me the .dat file and I used Bud's site to convert that to .csv. In that file I could see pitch and roll diverge at the same rate at the same time....which also matches the video in post 383....

 

[III% Streve]

Ha Ha! It proves absolutely that a prop will fly off with loss of one phase on the motor. I am anxiously awaiting your "proof" that active braking can easily cause this, and if it can, it's a poor design.

Its nearly impossible to safely CREATE an active breaking event. The only thing I will do is show that a rapid slowing of the motor will dislodge a prop that isnt properly tightened.

 

[TheRealNick]

Its nearly impossible to safely CREATE an active breaking event. The only thing I will do is show that a rapid slowing of the motor will dislodge a prop that isnt properly tightened.

Streve,

What about my idea of holding the landing skid while it is in hover and rocking it about? That seems very close to this situation....could be holding the phantom the whole time.

 

[flyNfrank]

Edit: TheRealNick

Wait second.....hold up. Now there has to be a reason your account just started 2 days ago. If you are posting with more then one account, which you just about have to be, then it is not good at all. There is no new person going to join in such a conversation like you have. I have never tolerated anyone to have 2nd accounts on discussion forums of mine, and yes this one is not mine. Just not good. If you want to send me a conversation/PM to clear this up then so be it. Otherwise you're going to be exposed.

 

[III% Streve]

View attachment 35852

In the above picture I paused the playback right at the start of the decent. You can see in the middle picture the phantom banked hard right. In the left picture you can see the phantom nose went down. This matches loss of lift on the front right corner, which matches the corner that had the only intact prop found detached.

From the video in post 140 it is pretty hard to make the props come off and McVentures said he did finger tight. I think if the torque was that critical we would see this all the time. I think something caused that motor to stop. Maybe not motor failure, a logic or esc fault perhaps, who knows...

Not trying to keep poking a dead horse, but amperage draw remained consistent to normal flight untill recording stopped, showing that 4 motors were still active. A shorted motor would cause a spike. A dead motor would show a drop. We don't see either characteristic. We all make mistakes from time to time. I flew to a site 3 days ago and forgot my **** SDCards! Is it possible that a prop want tightened by accident.. Sure. It sucks!! A LOT! But its a possibility. And all data points that direction.

 

[III% Streve]

Here is the data from the flight

 

[TheRealNick]

Not trying to keep poking a dead horse, but amperage draw remained consistent to normal flight untill recording stopped, showing that 4 motors were still active. A shorted motor would cause a spike. A dead motor would show a drop. We don't see either characteristic. We all make mistakes from time to time. I flew to a site 3 days ago and forgot my **** SDCards! Is it possible that a prop want tightened by accident.. Sure. It sucks!! A LOT! But its a possibility. And all data points that direction.

Tend to agree but by the same logic if the prop did come off the amperage should also have dropped since torque would be nothing. Only thing I can think of is maybe current capture is not that precise or filtered....

 

[III% Streve]

Tend to agree but by the same logic if the prop did come off the amperage should also have dropped since torque would be nothing. Only thing I can think of is maybe current capture is not that precise or filtered....

I was on that same path. It would be extremely telling if DJI gave us RPM! But at the same time, the trends in RPM at hoover would be impossible to predict. Much less during an uncontrolled descent!

 

[RobVson]

Call DJI. I had a loss of power crash a month after I bought my p3a, and they sent me a brand new one. It took 2 and a half weeks. They did not tell me what was wrong or what happened, so obviously they found the error in my favor. They were also super nice. I know other people have had different experiences, but that was mine. Good luck