Yet another flyway in the north

[ukaleq]

Hi fellow Yuneec users,

After a flyaway in Kangerlussuaq, Greenland last summer, I just experienced another one a couple of days ago in Ruka, Finland.
Ironically I had just received a replacement for the lost drone (Yuneec refused to acknowledged any responsibility and offered me a refurbished unit at a discounted price).
I am in no doubt that these drones are totally unreliable at those latitudes. I would appreciate your opinion based on the log files.

 

[DoomMeister]

Could you upload the Remote_GPS files also. The info I'm seeing in the warnings in the Sensor files say the Controller had insufficient GPS to record Home Position so when RTH was activated it continued to fly in the direction it was already headed. If you had returned to Angle mode with GPS instead of Manual mode (no GPS) that you were in when RTH was activated you may have been able to regain control. Definitely need the Remote_GPS files to better understand what exactly was going on.

All other flights seemed to go just fine. The test flight in Germany DID NOT test the functionality of the RTH. Is the ST-16S the same one from the first TH Plus?

 

[Steve Carr]

I am in no doubt that these drones are totally unreliable at those latitudes.

Seriously? Why would latitude affect the aircraft?

 

[Peggy]

Two flyaways?? **** that’s unlucky. I am assuming you did all three calibrations before attempting a flight? Then waited to receive at least 10 sats on both st16s and aircraft. 10 is the minimum you should take off with.

 

[Steve Carr]

Two flyaways?? **** that’s unlucky. I am assuming you did all three calibrations before attempting a flight? Then waited to receive at least 10 sats on both st16s and aircraft. 10 is the minimum you should take off with.

But.........15 is much better........and safer in my opinion. Especially if you have had problems in the past.

the Controller had insufficient GPS to record Home Position so when RTH was activated it continued to fly in the direction it was already headed

Which file is the flight in question???

 

[DoomMeister]

00018. He had 19+ satellites on the TH+, but the sensor file 032 had warnings about insufficient GPS in the remote and home position was not recorded. That is why I want to see the Remote_GPS files too. It is the only way to verify/debunk the warnings in the Sensor file.

@Steve Carr , do you have to enter Manual mode (Angle no GPS) with a soft switch like on the TH? Or can it happen automatically if the remote has too few satellites? That was the mode the flight was in when RTH was activated, and the aircraft continued along the path it was on just prior to RTH activation.

 

[10-8]

00018. He had 19+ satellites on the TH+, but the sensor file 032 had warnings about insufficient GPS in the remote and home position was not recorded. That is why I want to see the Remote_GPS files too. It is the only way to verify/debunk the warnings in the Sensor file.

@Steve Carr , do you have to enter Manual mode (Angle no GPS) with a soft switch like on the TH? Or can it happen automatically if the remote has too few satellites? That was the mode the flight was in when RTH was activated, and the aircraft continued along the path it was on just prior to RTH activation.

GPS signal on the remote is irrelevant and not used unless you are using the return to controller option. Home position as it relates to RTL is recorded by the 520 at the point of arming. If taken off manually before there is an accurate HP then the RTL point could be out of site when RTL is used. Easily avoided by going to manual of Angle if you see it going away from you.

The GPS in the controller is used for the distance display, return to controller, and loading the local maps in WP planning. It is not used in flight unless the return to controller is active, and then will upload a new HP to the 520 as you move. In testing, I will often fly with 0 sats on the remote as I am usually under a metal roof. Those flights usually conclude with a RTL.

 

[DoomMeister]

10-8, this was a TH Plus not an H520. The warnings in the Sensor file stated the Home point was not recorded.

 

[10-8]

10-8, this was a TH Plus not an H520. The warnings in the Sensor file stated the Home point was not recorded.

Thanks, I always check both forums and forgot which one I was in. But same result. HP is recorded at time of arming by the + GPS. FYI.. I fly the + under the same metal roof..

 

[ukaleq]

Here is the complete log. How does one see which mode has been switched by the user ?

I have always flown in angle mode and, in retrospect, I should never have trusted it given my past experience and the instability encountered in previous flights, and I should have flown instead in manual mode. The thing is, even though it was materially impossible to fly in manual mode when I had my first fly away (because it was not included in the firmware at the time!), that didn't bother Yuneec, who said I had deactivated the GPS. This time it would indeed have been a possibility, which sadly I didn't use...

The rear LEDs were solid purple when I took off, which according to the manual means "angle mode with GPS fix".

Whether the home point was recorded or not, there is no justification whatsoever for the drone flying in the wrong direction with no stick input (at a speed that is well above what could be ascribed to wind drift).

@Steve Carr "Seriously? Why would latitude affect the aircraft? "

I do not know the exact explanation but both DJI and some PX4-based drones (at least with earlier firmwares) have been known to have had serious issues at high latitudes. I myself experienced plenty of GPS signal losses with a phantom 4 (which is ok) but also several episodes of total loss of control with the aircraft spiralling at full speed. It seems to be caused by some inconsistency between the GPS signal (which degrades at high latitude as the sats get "closer to the horizon" - although ca. 66°N should not be critical) and the compass data. In the case of the compass, it is the proximity with the magnetic pole, rather than with the geographic pole, that matters. The compass gives the direction of the horizontal component of the magnetic field but when you get closer to the magnetic pole, that component gets smaller and smaller and measurements tend to behave more and more erratically. Now, DJI explicitly writes in its manuals that its aircrafts do not work in "polar areas" (which is of course meaningless without a proper definition of 'polar areas') but Yuneec doesn't.

 

[ukaleq]

How can the home point not be recorded when the logs show that both the aircraft and remote stored the right GPS coordinates?

 

[DoomMeister]

@ukaleq,

Thanks for uploading the logs including the Remote_GPS. Looking at that matching file shows that it had 16 to 19 satellites at all times, so the warnings from the PX4 Sensor file about Weak Remote GPS signal makes absolutely no sense.



This is a plot showing Flight mode vs Aircraft satellites vs Remote satellites. Flight mode shows you starting in Angle mode w/GPS (mode 5), then changes to Manual mode 4 (Angle wo/GPS), then near the end RTH landing (mode 13). From your description you were flying in Angle mode with GPS at takeoff and the system dropped into Manual mode on its own due to perceived lack of GPS signal (based on Sensor file readout).

I am attaching the AppLog from reading Sensor00032.txt for your perusal. I will return to this later with more info, I have an appointment to go to.

 

[DoomMeister]

This is a comparison of the telemetry of the last flight from the perspective of Sensor data and from telemetry files.

telemetry from the sensor file ends here which comes from the PX4.


Note the similarities to the Sensor telemetry at this point. This is where RTH is initiated and there is no further telemetry from the Sensor file.


This is the path after initiating RTH.

Yaw inputs were reflected in small deviations left and right of main direction of flight and may have had some effect if held long enough in one direction to effect a change in flight path. Yet the initiation of RTH should have brought the aircraft closer not send it off heading for the Far East. My main concern is the abrupt end of the Sensor file when RTH was initiated, and the RTH function was not tested at the repair center in Germany.

 

[AH-1G]

@DoomMeister
Doom, If your correct and it headed east> it went approximately .48 of mile over the hill past a RED and White stack.
landed approximately 66°16' 48.2" North 29°15' 65.12 E Using Google Map

 

[AH-1G]

Google Earth and Google Map show two totally different Longitude and Latitude for the same location at crash area.?
Google Earth 66°09'53.14" N 29°09'22.16" E
Google Map 66°16' 48.2" N 29°15' 65.12" E

 

[DoomMeister]

It did not land there, it was at 110 meters above takeoff point and headed east at 13 m/s (almost 47 kmh or 29 mph) and 15.9V left on the battery. The telemetry ends there and there has to be a good 7 minutes or so of flight time left before the first low battery warning.

 

[AH-1G]

It did not land there, it was at 110 meters above takeoff point and headed east at 13 m/s (almost 47 kmh or 29 mph) and 15.9V left on the battery. The telemetry ends there and there has to be a good 7 minutes or so of flight time left before the first low battery warning.

4.83 miles?

 

[AH-1G]

So it may have headed towards Marttila or Valitalo.

 

[AH-1G]

It did not land there, it was at 110 meters above takeoff point and headed east at 13 m/s (almost 47 kmh or 29 mph) and 15.9V left on the battery. The telemetry ends there and there has to be a good 7 minutes or so of flight time left before the first low battery warning.

How do we know if all of the sudden it lost power?

 

[AH-1G]

Once he went over the hill no telemetry can be recognized, correct?
It should have come back home?