endotherm
 First Officer
Flight distance : 503241 ft
Australia
Offline
|
Looking at all the flight logs gives very good data describing exactly what we are seeing in the video in your first post. There is nothing really out of place or remarkable about it. Although you see the wind at 30km/h (~8m/s) at ground level, it increases as you climb. It is invisible, so there is no way of knowing just by looking, which is why we are relying on telemetry. FYI, the Phantom 3 top speed is around 15m/s (55km/h) and when in RTH, it is only around 10m/s (35km/h). What is really striking to me is the second video you posted. It shows you ascended at about 15km/h, then you noted yourself it was being picked up by the wind at around 200m, at about 20km/h horizontally. Again at 300m it was 30km/h, 400m was 40km/h, 500m was 50km/h (fluctuating).
Once you got to 500m, you turned the aircraft to face you, only to watch the aircraft moving away from you. Strangely there is only a single timid movement of the right stick forward (towards home) at 50-75% for a second or two (lets say 10 seconds in real time, as the video is sped up 8x). Nothing aft, or left or right, yet you conclude the stick has somehow failed. The onscreen joystick movements are validated with the figures in the flight data. Hardly any movement.
You then hit RTH in a panic, and wisely left it to do its thing after again rotating into its direction of travel. You observed this wasn't helping. You decided to reboot your controller (hey, why not?). After it resumes, we watch it drifting away in excess of 50km/h. You then describe an "auto-RTH"(?) kicking in which again rotates the aircraft, correctly showing no manual stick input, but it again faces you. Nothing recorded in the log messages to show auto-RTH. At this point the aircraft is hitting in excess of 20m/s (75km/h). Again a couple of timid movements in both sticks, then a nice aggressive 100% forward movement for 24 seconds, followed by another for 48 seconds, then back to idle. Those moves may have just started to decelerate the aircraft from 50km/h (we observe the speed drop to the high 40's momentarily), but were nowhere near long enough to make a dent on its speed. Then another 16 second burst, a decrease in altitude to 430m where the speed has now dropped to ~40km/h. Somehow you think 430m @2.25km is too low and you might lose signal if you go lower? I really doubt that. Then there is another minute of 100% forward, which manages to hold the aircraft at 42km/h. Once you let go the speed picks up again to 55km/h (13km/h was the effect of your acceleration against the wind, which you just abandoned). Later, another 40 second burst of speed then you just let the sticks go and watched it drift towards the lake for a minute and a half.
You had 4 minutes 15 seconds until a critical battery forced it to land. Based on an airspeed of around 50km/h on a heading of 025° I estimate it got carried 3.54km before it started descending. Descent is limited to 3m/s (10km/h) to prevent vortex ring state and a crash from turbulent air while descending. From 433m it would take around 144 seconds to land. Let's allow for reducing windspeed at lower altitudes, so half of the trip at 40km/h, and the rest at 30km/h down to the ground where you already measured it at 30km/h. 72 seconds for each (1/50 of an hour) = 0.8km + 0.6km, or 1.4km during the descent. If we look at where we lost telemetry and extrapolate from there, our best guess is that it landed 4.94km further along, give or take due to fluctuating winds and other assumptions. If you are lucky it might be recovered one day washed up on the shore.
extrapolated landing
Unfortunately all the evidence points to the crash being attributed to strong winds at altitude and an inability of the aircraft to overcome the opposing forces and fly in those conditions. Other factors include piloting errors such as not descending, unfamiliarity with conditions at altitude, and just allowing it to be taken with the wind without changing course or applying continuous forward thrust. Hopefully you can learn from this, and especially reconsider flying at such altitudes in future.
In response to your earlier points:
1 At no point in the fpv video did I see the usual props in the frame - this would be expected, especially when flying into the wind.
At all times when flying into the wind, the aircraft was tilted at ~22° towards the home point (arms closest to home were lowered). This was an automatic attempt to hover and hold position. Coincidentally, the camera is reported tilted down at -21.4° for much of the flight, except when facing down or traversing. This is good framing which shows the ground and avoids large chunks of sky. Consequently the camera view was low enough when facing towards home point to miss the movement of the arms and spinning props, and more so when facing away as the arms were even higher.
2 My speed increased and decreased throughout the flight, just like you would expect wind to do.
This is a characteristic of the wind. The speed fluctuations were not noticeably a result of any power you were applying, only after sustained periods of countering the wind did we see any lowering of the speed that was consistent with, or may have been attributable to aircraft power.
3 When I hit RTH, even in the approx. 50km/h wind at this height, my speed should have at least reached nearly zero.
Inertia plays a big part here. It would take some time with sustained power to even approach zero. With fluctuating winds it is impossible to tell. In RTH mode your top speed is only 35km/h.
4 I had flown on a day with the same wind speed just before to 400M, and was able to fly into the wind at a reasonable speed.
Wind is unpredictable, what it was doing minutes before is no indication of what it will do in the future. Unless you are up there with an anemometer you are just guessing.
5 The Phantom was holding inside 1M (incredible, right), this would not be expected if it was about to be blown away. At around 200M high, a steady and increasing drift began, more in line with the Phantom not trying to hold position.
At ground level you measured 30km/h winds. The aircraft is capable of 55km/h when flown manually. It is capable of overcoming those forces and remain stable.
6 There are videos that show the Phantom in extremely strong wind, and it whips about wildly to keep its position. Mine was a gradual drift that increased as the phantom built momentum.
They could have been buffeting winds changing intensity and direction wildly, This seemed more steady in direction and reasonably constant, with gusts up to 75km/h as measured by the aircraft flight data recorder and displayed on your video. At those altitudes there is little reference to anything that would indicate the aircraft whipping about. I was able to see that the aircraft was moving with some speed by watching the roads quickly disappearing in the picture beneath.
|
|
lvl.1
Core User of DJI
Captain
lvl.3
lvl.4
DJI team
Second Officer
lvl.2
