Looks like TBE = bad calibration. Always double check compass mod after calibration and before every flight . Hopefully you can get back up in the air soon!! Good luck !!
First flight fly away....P3A crash...
- Thread starter TmoneyF
- Start date
Are you going to post the full video then? I find the lack of video from close to the powerlines not very coincidental.
1. I don't have the full video from the SD card, It stops just before the copter started to drift without any inputs and me going into panic mode. The video of where things when wrong is here on the site, It's was pulled from my phone.
2. You obviously haven't read the entire thread.
3. Why would I have a reason to lie? Who on the phantom pilots message board am I trying to impress or pull one over on? The dealer said they would replace the copter anyway I just posted here to help contribute to the community.
@TheRealNick, @TmoneyF There is an issue when the internal representation of the orientation is converted to the roll, pitch, and yaw numbers that you see. It's called gimbal lock if you're interested. It's apparent in the pitch where the value jumps from -90 to +90. That's physically impossible. The same happens with yaw and roll.
Bottom line is I don't think these numbers can be used to determine if the P3 was upside down in this case. The accelerometer data might give a clearer picture
@BudWalker. I wasn't using the pitch, just looking at the mag, it should swap between the axis as you rotate it. Whatever it was for part Z is what it should be for part X for part 2. I used my Phantom last night to determine the x axis is front/back, the y axis is left/right, and the z axis is up/down. I checked the acceleration data in @TmoneyF and I am pretty sure he had it with the side down for part 2. I am not sure what this would do, especially since the yaw and orientation were correct in the playback, but I wouldn't do it again. I thought about trying it with mine, but don't want to take that risk. I would if I knew the Phantom ignored compass in ATTI mode but I am not sure...Calibrate level and then with the nose down.
Also @TmoneyF, I think the earlier post about the video being cut was inferring the powerlines interfered with the Phantom...i.e. they made the video cut our rather than you trying to lie or something.
Also @TmoneyF, I think the earlier post about the video being cut was inferring the powerlines interfered with the Phantom...i.e. they made the video cut our rather than you trying to lie or something.
To check how a calibration looks like on the 3 Phantom Gauges in Dashware I taped myself doing a calibration procedure (1st 30secs in the video). Then did some wild stuff. Then synched the Dashware render to the video.
With FLY013.DAT I rendered the calibration procedure starting and ending as per the ticks of the log file.
Step 1 of calibration, the anticlockwise yaw is correct. Then there is a 20 second interval before step 2 is performed. And instead of pitch being applied its clearly roll.
Why would there be roll? Is it due to a magnetic source or is it an incorrect manipulation? Whichever reason, could this be the reason of the drift? Anyone want to try calibrating using roll instead of pitch and see whether calibration reports as successful (be warned this could lead to a crash)?
With FLY013.DAT I rendered the calibration procedure starting and ending as per the ticks of the log file.
Step 1 of calibration, the anticlockwise yaw is correct. Then there is a 20 second interval before step 2 is performed. And instead of pitch being applied its clearly roll.
Why would there be roll? Is it due to a magnetic source or is it an incorrect manipulation? Whichever reason, could this be the reason of the drift? Anyone want to try calibrating using roll instead of pitch and see whether calibration reports as successful (be warned this could lead to a crash)?
Nice video and dashware, all your hard work helps everyone understand! Also, perfect compass calibration!
I am pretty sure it was calibrated with roll (side downward) instead of pitch (front downward), you can check the accelerometers and the y goes to -1 during the second part of the calibration, I have no idea what this would do....graph below...you can see Z accelerometer goes from -1 to 0 and Y accelerometer goes from 0 to -1 and Mag Y goes to value Z was before (I took Mag Z off for clarity, but it is in an earlier chart I posted)....
I agree that wrong or bad compass calibration caused this incident. I think its a good idea to make sure that before every flight... make sure everything is done correctly and in the best condition. Its like real pilot, before flight...they have to make sure theres no error...or...something unexpected waiting ahead...
Hi luap, what app do you use to analyze and get such interesting reasult. Can you share?
My take on that first video is it got close to the power lines and the P3 freaked out, they put out a lot of juice, and the magnetic field on them power lines is quite a spread out..
My take on that first video is it got close to the power lines and the P3 freaked out, they put out a lot of juice, and the magnetic field on them power lines is quite a spread out..
I fly within feet of power lines nearly every single flight, for over 100 flights... never had an issue. I really do not think power lines will cause an issue for a phantom, unless you hit them, additionally, on the way to taking my phantom out nearly every flight i have to fly over these because i get the best distance in that direction: see photo below. Never had an issue.
Last edited:
I looked at the accelerometer data and it does indicate that the 2nd step of the calibration was done incorrectly. I.e. instead of nose down it was done with the right side down. It's possible that this could have resulted in a bad compass calibration and subsequent incident. Some would say it's obvious. Although I don't have a better explanation I'm not convinced. The most I can say is I don't know what the cause was.
Contrary to popular belief the compass calibration is not about discovering which way is North. Rather, it's purpose is to adjust the gain of each of the X, Y, and Z magnetometers so that they have a uniform response to the geomagnetic field. All that's really required is that the P3 be oriented in enough different directions so that each magnetometer is exposed to a range of values.The first step of the compass dance gathers that data for the X, and Y magnetometers. The second step has the Z axis pointing at the horizon so that the data can be gathered for it.It probably doesn't matter that it's the X axis pointing at the horizon instead of the Y axis. I.e., right side down should work just as well as nose down for the 2nd step.
I once did an experiment where the compass dance was done upside down. 1st step top down, and 2nd step nose up instead of down. I had carefully measured the X, Y, Z, MOD numbers before and and after and didn't see any difference. Flew fine.
This flight was fine until the compass data started showing interference. That doesn't come from a bad compass calibration it comes from some external input. It's possible the invalid compass data didn't have anything to do with the loss of flight control. Recall that there was a loss of video implying that other functions could have also been lost.
Contrary to popular belief the compass calibration is not about discovering which way is North. Rather, it's purpose is to adjust the gain of each of the X, Y, and Z magnetometers so that they have a uniform response to the geomagnetic field. All that's really required is that the P3 be oriented in enough different directions so that each magnetometer is exposed to a range of values.The first step of the compass dance gathers that data for the X, and Y magnetometers. The second step has the Z axis pointing at the horizon so that the data can be gathered for it.It probably doesn't matter that it's the X axis pointing at the horizon instead of the Y axis. I.e., right side down should work just as well as nose down for the 2nd step.
I once did an experiment where the compass dance was done upside down. 1st step top down, and 2nd step nose up instead of down. I had carefully measured the X, Y, Z, MOD numbers before and and after and didn't see any difference. Flew fine.
This flight was fine until the compass data started showing interference. That doesn't come from a bad compass calibration it comes from some external input. It's possible the invalid compass data didn't have anything to do with the loss of flight control. Recall that there was a loss of video implying that other functions could have also been lost.
@TmoneyF - Did you ever get the video off the SD card? I would suspect the video is not cut in two like the videos you got off the transmitter. Sitting here thinking about it, perhaps the RC signal was lost which caused the video to split and the Phantom was not quite yet in Return to Home (RTH) but also not holding....that would match the 3 second timeframe of the drift...
Anyone know exactly what happens during the 3 seconds before a Phantom starts RTH? Does it stop? Does it drift? I know a lot of people fly it out until it loses signal and then it RTH, does it keep going forward for those 3 seconds?
Anyone know exactly what happens during the 3 seconds before a Phantom starts RTH? Does it stop? Does it drift? I know a lot of people fly it out until it loses signal and then it RTH, does it keep going forward for those 3 seconds?
yeah thinking more about it, even if it lost signal it should have never violated the distance...
One would think that if the compass mod is not in the acceptable range or if something isn't right with the compass that the phantom would not to allowed to take off until it is corrected . Sorry for the run on sentence.
I was able to get the video off the SD card there was only one file and it stops just before the infamous "drift" begins.
Sent from my iPhone using PhantomPilots mobile app
Those high-voltage, high-power transmission power lines have to be highly suspect.
They could both be emitting a lot of RF interference, and also a strong magnetic field. I think you need to find a new place, way out in the open, to try your initial training flights.
Just my best guess. Good luck.
They could both be emitting a lot of RF interference, and also a strong magnetic field. I think you need to find a new place, way out in the open, to try your initial training flights.
Just my best guess. Good luck.
Similar threads
