Hello Fellow Yuneec Pilot!
Join our free Yuneec community and remove this annoying banner!
Sign up

H520e dropped out of sky after "No. 4 motor failure" error

Joined
Jan 31, 2022
Messages
3
Reaction score
1
Age
48
Over the weekend I flew a standard survey mission. About four or five minutes into the flight, I got a "no.4 motor failure" error and then a "vehicle will return to take off area". This did not happen. Instead, the drone literally dropped out of the sky. There is damage to both the drone and the E90. Interestingly, none pf the props were damaged which surprises me greatly. I had the altitude set at 100' as the area being surveyed is treed. I am having trouble contacting Yuneec as their phone number just gets a busy signal and I don't have an email to send to other than the submission form on their website. I purchased this drone directly from Yuneec US website. Any advice appreciated.
 
Keep trying the phone number between 9 AM and 5PM Pacific Time.

In the meantime you can download the ulog (.ulg) files from the aircraft using QGroundControl as shown in this video (H520 and H+ use the same procedure).


Once that is done upload the ulog file for the failed flight to PX4 Flight Review Flight Review and copy the link to the resulting analysis and post it here. We may be able to determine the cause of the failure while you are waiting on Yuneec USA.
 
  • Like
Reactions: 7cyclops
I just had a look at your data. I cannot see any obvious fault with you drone so. Also the flight went exactly according to plan until that very last moment when it started to spin. Interestingly that happened in the very same moment when you arrived back at some higher trees. Precisely at the location where it started to spin there is a tree which ist slightly higher then the ones surrounding it. Maybe it is just a coincidence but I can see a chance that you might have touched it. Unfortunately I cannot measure its height since the available data on the internet is insufficient.

What also supports my assumption is the fact that the terrain has a slight rise towards your crash site. You started at a an altitude of about 99m (324ft) above sea level. The crash site however is at an altitude of about 110m (360ft). Your flight attitude was like you said about 33m (100ft) above ground when you started. Since the drone is only measuring barometric height and therefore is maintaining the same flight level of about 122m (400ft) for the entire flight, your altitude was only about 12m (39ft) above ground just before the crash. That altitude might have been lower than the surrounding trees. At least it was fairly close and I would not recommend to take a risk like that.

What is your estimation towards the height of the surrounding trees at the crash site?
 
I suspect @jannislh is correct from observing the flight in 3D view. Another possibility is a bird strike and at that altitude recovery in 5 rotor mode would be unlikely. Notice direction of flight veering to the right after failure of motor 4 (right rear motor) causing aircraft to tilt to the right.
 
I also suspected a tree strike initially. The back part of the property has the youngest (and shortest) trees as it was mowed down completely about five years ago. Estimate of tree height in the back is 25-30ft.

What made me question a tree strike was that there is no prop damage at all. I’ve flown many craft, from 35% scale planes to 25% helis. Typically if a blade or prop hits a tree, you know it.

The other thing that made me question a tree strike is that just before the crash, the video started lagging badly and even had a black screen for a split second. I was not sure if there could have been a connection problem due to the density of the trees.
 
What made me question a tree strike was that there is no prop damage at all. I’ve flown many craft, from 35% scale planes to 25% helis. Typically if a blade or prop hits a tree, you know it.
I totally understand your idea but what you have to mind is the rotary mass and the stiffness. The moving parts on the H520 are fairly lightweight and the Props are very flexible. I never had an OEM prop fail on my Typhoon H. Not even when it tipped when I landed on a rough and hard ground during windy conditions.

When flying my FPV racing drones I noticed that even a slight contact has severe effects on the flying characteristics. Especially when you are flying slow a contact with thin leaves and branches will already pull you towards the tree since the propellers on that side are slowed down. When you are going like 80mph however, the same leaves and branches will hardly slow you down at all. Even though the props are spinning at more than 40.000rpm they will only brake on very rare occasions. Usually only the leading edge gets eaten away until you start to notice an increase in noise and vibration.

In summary I am not surprised that your propellers are not broken.

The other thing that made me question a tree strike is that just before the crash, the video started lagging badly and even had a black screen for a split second. I was not sure if there could have been a connection problem due to the density of the trees.
Since you were flying low in a distance of about 316m (just below 1000ft) without a direct line of sight, I guess that was related to a poor video signal. Unfortunately there is now useful RSSI reading in the log so I cannot give a clear statement. But what you can see is that the signal failed entirely just 4s after the motor 4 error was noticed.

What I also found interesting is that motor 4 was indeed the first motor to notice an issue but not the last. Within the very same second all six motors recorded the same error. That also supports my thesis that the drone collided with a tree since it indicated that none of the rotors was able to spin freely while the drone was still falling towards the ground. Also the fall took a lot longer than you would expect for a free fall. The drone started falling after around 3:05.775 of flight time and touched the ground at 3:08.580. Almost 3s is very long for a fall of only 12m (39ft). Therefore it is very likely that the fall was slowed down by something. The drone itself has most likely only contributed very little since it was spinnging around all axis and was even upside down for a moment.

In conclusion I am almost 100% sure that it was a controlled flight into one of those trees. All the data collected show perfect health of the aircraft itself but do show a lot of indications towards a collision.
 
  • Like
Reactions: veearnis

New Posts

Members online

No members online now.

Forum statistics

Threads
20,952
Messages
241,578
Members
27,284
Latest member
csandoval