Almost lost my p3a due to error

[Fly Dawg]

I would try 2 things first, before looking in to any possible hardware issues. First, find a place nearby in a WIDE OPEN area with nothing within 100 yards or so of you and do a compass calibration there. That way you should be able to eliminate any possible calibration errors that could have been introduced. Then try that same flight again. ( Or close to it ) and see if you get any messages. If not that should correct the issue. If you still get the errors, then try a cold IMU calibration. In the interim, you could pull the .dat file from the AC and place a link from a sharable location back here so we can view that file. It may or may not offer more clues as to what may be going on. If you have not done that before, see the link below.

How to retrieve a .DAT

 

[randyvanscoy]

I would try 2 things first, before looking in to any possible hardware issues. First, find a place nearby in a WIDE OPEN area with nothing within 100 yards or so of you and do a compass calibration there. That way you should be able to eliminate any possible calibration errors that could have been introduced. Then try that same flight again. ( Or close to it ) and see if you get any messages. If not that should correct the issue. If you still get the errors, then try a cold IMU calibration. In the interim, you could pull the .dat file from the AC and place a link from a sharable location back here so we can view that file. It may or may not offer more clues as to what may be going on. If you have not done that before, see the link below.

How to retrieve a .DAT

Ok i just watched a urube on doing a cold imu calabration. I have done them before but not COLD. Could compuse be intermitant like a pinched wire or bad unit? Due to it flying ok for a while then went crazy? Just scared to death to fly again kinda.....!!!!!! Thanks for info. I will try this afternoon . Wind is 0 on ground level amd about 60 degrees outside

 

[Russ43Phantom]

I had that experience only once. It was when I launched from a park picnic table that had a metal frame supporting a wood table top. That little amount of metal caused compass problems and I had to bring the drone down into grass from 5’ high. The P3 was totally out of control. I learned to carefully look for any metal in the takeoff area and avoid it.

 

[randyvanscoy]

Ok i just done the
Following.
Placed drone outside in cool air (45 degrees) for 2 hours. Started up go app and got all connected to drone. Turned on battery for drone and went to calabrate the imu. It went through ok. I then took out in the moddle of a field with nothing around and calabrated the compus. It was ok after. Now i went on and started up and let it hover for a while. Seemed ok with nothing unisual happning. So i flew a little higher and ok so far. Decided that all was looking good. Wrong! Inwent to fly up and away from me for a short distance, when i got a compuse error and it started to take off on its own. I got it down to about 10 feet and shut it down by the sticks methode and it fell and landed upside down in grass. It is fine shape, but dont know where to go with it now.

 

[Fly Dawg]

It is fine shape, but dont know where to go with it now.

As I mentioned earlier, pull the .dat from the AC, per the instructions and place a sharable link back here. There are many of us who can take a look and see what the issue might possibly be. That would be the next step, before you move on to a hardware issue with the AC. And when you you say shut down by the "Sticks" method, if you were using CSC, that is why it flipped over. NEVER EVER use CSC to shut down the AC. Simply pull the throttle down for a few seconds. That is the way to do that.

 

[randyvanscoy]

here is the text file, the data file when i try to upload it here, i get a error file is to big.

 

[randyvanscoy]

Airdata UAV - Flight Data Analysis for Drones

 

[randyvanscoy]

I would try 2 things first, before looking in to any possible hardware issues. First, find a place nearby in a WIDE OPEN area with nothing within 100 yards or so of you and do a compass calibration there. That way you should be able to eliminate any possible calibration errors that could have been introduced. Then try that same flight again. ( Or close to it ) and see if you get any messages. If not that should correct the issue. If you still get the errors, then try a cold IMU calibration. In the interim, you could pull the .dat file from the AC and place a link from a sharable location back here so we can view that file. It may or may not offer more clues as to what may be going on. If you have not done that before, see the link below.

How to retrieve a .DAT

when i try to send the .dat file, i get a error (file too big). what do i do now. i downloaded the data files from bird and stored all on a flash drive.

 

[Fly Dawg]

what do i do now. i downloaded the data files from bird and stored all on a flash drive.

You cant upload those directly. What you need to do is upload the .dat file to a "shareable" location, such as Drop Box, Google Drive, or a similar area where you can share the download. And place a link back here to access the file.

 

[randyvanscoy]

Dropbox - FLY465.DAT

this is the share of the first flight data file i hope.

 

[randyvanscoy]

Dropbox - FLY468.DAT
this is the second flight.

 

[randyvanscoy]

As I mentioned earlier, pull the .dat from the AC, per the instructions and place a sharable link back here. There are many of us who can take a look and see what the issue might possibly be. That would be the next step, before you move on to a hardware issue with the AC. And when you you say shut down by the "Sticks" method, if you were using CSC, that is why it flipped over. NEVER EVER use CSC to shut down the AC. Simply pull the throttle down for a few seconds. That is the way to do that.

i would of done the proper way to shut down, but it was heading straight for a fence at high speed run away. didn't have time i thought to wait . luckily it didn't do any damage to it.

 

[Fly Dawg]

Ok, took a quick look at the second flight data. I am suspecting that you may have a bad compass, but am going to ask @sar104 , @BudWalker a couple of folks to have a look in the interim. Am in and out here today, so can't go much more than just the quick glance at the moment.

 

[randyvanscoy]

Ok, took a quick look at the second flight data. I am suspecting that you may have a bad compass, but am going to ask @sar104 , @BudWalker a couple of folks to have a look in the interim. Am in and out here today, so can't go much more than just the quick glance at the moment.

sweet....thanks. i just need some ideas...hope i got the right files downloaded to help. there is two flights of data loaded.

 

[Fly Dawg]

Yes, the second one is the one I took a quick look at, being that it had the same errors as the first.

 

[BudWalker]

@randyvanscoy , @Fly Dawg

FLY465 appears to be the continuation of FLY464. The motors didn't run and it looks like the P3 was sitting around until about 1000 secs when it was moved some. There are also repeated CSCs about 1000 secs but no motorStart because of a battery low
[M.Start]REQ_RC_NORMAL FAIL, battery user low land level stop motor!

The dateTime for FLY465 was 2018-3-4 0:15:52 GMT which is 23 minutes after the first .txt submitted. FLY465.starts at the 16 minute mark making it likely that FLY464 isn't the right .DAT either. I suggest that you submit FLY463 and FLY464.

FLY468 wasn't an actual flight - there wasn't a motorStart and the control sticks weren't moved.

 

[Fly Dawg]

@BudWalker As I said this was a real quick look and I negated to check for motor start ( My bad). I was looking at these two parameters in the CSV. And the values were all over the place and skipping in FLY 468.

directionOfTravel:mag[degrees [-180;180]]
directionOfTravel:true[degrees [-180;180]]

 

[BudWalker]

@BudWalker As I said this was a real quick look and I negated to check for motor start ( My bad). I was looking at these two parameters in the CSV. And the values were all over the place and skipping in FLY 468.

directionOfTravel:mag[degrees [-180;180]]
directionOfTravel:true[degrees [-180;180]]

That's because it wasn't moving and what you're seeing is essentially a presentation of Longitude , Latitude noise.

 

[Fly Dawg]

That's because it wasn't moving and what you're seeing is essentially a presentation of Longitude , Latitude noise.

Well, that's what I get for a quick look....and makes sense now that you mention it.

 

[BudWalker]

@BudWalker .......I was looking at these two parameters in the CSV. And the values were all over the place and skipping in FLY 468.

directionOfTravel:mag[degrees [-180;180]]
directionOfTravel:true[degrees [-180;180]]

A while back another user encountered the same problem. To fix it a threshold was implemented that would require the AC to move a distance greater than the threshold for these two values to be valid. By adjusting that threshold your plot looks like this now.

Ideally, you'd want this plot to not show any valid values, but it's difficult to do this without invalidating data that really should be valid. The next version will have this new threshold.