Driffill wrote:How many surveys have been completed so far? When can we see some results to look for trends?
There's 155 responses as of today, and the rate has dropped off a bunch, so we seem to have gathered all we can.https://docs.google.com/a/deconcept.com ... wanalytics
I wrote up a summary on another phantom board, which I'll copy here:
I wrote this up when it was at 120 responses, looks like the data is about the same, but feel free to look for trends in the free text section at the bottom of the link above.
Of the 120 that responded, 32% have had a "flyaway" happen. Note that the definition of flyaway is pretty loose, and who knows if it was actually one or just inexperience, but that is still a really high number to me. Also note that it's pilots, not flights, and many people report flying most of the time with no problems... but still...
IOC was overwhelmingly in the "Off" mode - It would be interesting to see how often anyone actually uses those modes, as I never do myself, so this is probably irrelevant to the flyaway problem.
Also, 87% of flyaways happen while in GPS mode, but a small number (5 total, 13%) reported flyaways in ATTI mode - this could be within our margin of error or close to it.
32% reported regaining control of the copter after losing control. This is good.
Of those 32%, the top action that they did was "nothing, control came back automatically". A small number also reported switching GPS/ATTI mode, but most (61%) never regained control and the copter was lost or crashed.
55% of responses reported 25+ flights of their copter in its "current configuration."
61% reported an issue when flying in the "same location" as the previous flight. A hunch I had was that the GPS in the copter could be remembering a previous location as home and somehow activating RTH, but this question seems to have assuaged that thought.
Flying experience is a little interesting. 16% of the pilots with flyways reported that this is their first RC flying device, and only 16% reported being able to fly circles around indoors (ie very good). The majority reported being comfortable flying, so this could potentially rule out the problem being just "noobs".
The FPV question seems to rule out any interference with an FPV system. I was surprised to learn that 92% of pilots weren't using any FPV device at all, and the other 8% were using Fatshark 5.8Ghz (8% is within the margin of error here, so is likely negligible). So we can't blame interference from that.
Most people were using GoPro v3s, with half reporting the Black version, and 24% with no camera at all. So it seems like the camera (wifi interference) is probably not a problem.
66% of pilots reported doing an IMU calibration "0-5 flights" before the flyaway, so this could potentially rule out issues with the IMU getting out of whack after many flights. Perhaps in another survey we can dig into what happened just before the flyaway (ie a crash, or something like taking the copter on a plane or something).
66% also reported calibrating the compass within 0-5 flights of the flyaway, which seems pretty good to me and shows that it can happen even if you have a fully freshly calibrated copter.
Reading through the text descriptions I don't really see any patterns, but feel free to read through if you are interested.
I did send this to DJI and got a response saying "thanks" and that they would have a look at it, so perhaps it will help them narrow down the problem by eliminating some guesses.
Thanks all for the replies! And feel free to keep sending this around to other phantom pilots, more data is good!