Since I experienced several unexpected falls I decided to recreate the conditions in order to look for the possible reasons.
During the last days of 2021 I found a great location to fly my Anafi, just along a beatiful natural beach on the south-west of the Sardinia island.
The wind velocity is always above 5m/s at >30meters during winter, and mostly from north-west.
After 10 minutes of flight in Film mode where I was testing the ability to stay in a position without drifting away with the winds I changed to Sport mode.
Great wifi signal, 37meters AGL, speed was 12m/s with almost rear wind of 5m/s.
When it was 280m away I saw on the display a tipical free fall case: drone rotating on every axis falling down without control.
GPS position was recorded and the drone found in the bushes along the beach.
Il was emitting an intermittent sound from the motors that was determinant in order to find it.
I donwloaded all the flight log data and watched the video founding something very strange:
- I saw the fall on my phone, from sky to bushes and the video was visible for 2 seconds after it was on the ground BUT the video recorded was stopped at a point when the drone was in the sky and under control.
- the logs showed a partial change in the height record and battery: they looked truncated to 0 just before the fall.
Anafi flew many other times, even with strong winds, and I managed to send it >1km away over the sea to reach some little islands with no issues at all.
In the last week I mounted the modified battery (900mAh 7,6V 60C) and during a drain test (usually 9' 15" until auto landing) I assisted to a fall, inside my house.
Anafi banked left without control and fell, emitting the same intermittent saound from the motors.
The voltage was above 7'500mAh when rebooted.
Then removed the mainboard and double cheched every connector:
- the down facing camera connector has a little way to move on its side and slightly up near the mainboard
- the little flat power connector to the mainboard also could slightly move, even if kept in place by the plastic clip.
I decided to mount some plastic spacers to increase the pressure on those connectors, and did another try to find the problem.
Yesterday I managed another drain test with the light 900mAh battery, and thid time I got exactly what happened:
- during an hovering test i touched every drone arm and found some vibrations, that kind that could easily cause problems on the connectors! (the actual propellers are not the original ones, maybe they are not balanced and it's not an easy task with this split type)
- anafi did a safe autolanding at 9'15" (from 1 meter of height)
- I left the Anafi leaving on the carpet for about 20", with video recording (launched after landing) and the voltage had a very rapid drop down to <5000mV
I was looking the phone, video was still ok during the drop and wifi was working.
Few seconds later the drone entered in the same state I heard after the crashes: intermittent sound from the motors!
I put another battery and downloaded the videos, so the logs.
That's the point:
- when the battery have a drop in voltage WIFI can work and send video from camera BUT the recording stops
- the Logs also are stopped before the drop, and NO CRASH are reported (just like the other two times!).
MY CONCLUSIONS:
If the Anafi fall from the sky, sending video to your phone until down to the ground, BUT the video recorded stops just before the fall, then it's a battery problem related to a current/tension drop. I had proof that Anafi can send video with voltages below 5000mV, wich is not enough to record the crash in the logs or to write in the SD card.
If it falls without sending video to the wifi then the problem could be connector related (power to mainboard due to vibrations).
In the past I had several messages, in FreeFlight, about some 'vibrations'. They appeared just when I used the non original props.
1) Check your batteries for the real capacity and flight time
2) Use only original propellers and check for any kind of vibrations: even if you don't see them on the camera they can compromise the security of your flights!
3) Read the logs, but don't expect to find any data recorded under 6000mV.
During the last days of 2021 I found a great location to fly my Anafi, just along a beatiful natural beach on the south-west of the Sardinia island.
The wind velocity is always above 5m/s at >30meters during winter, and mostly from north-west.
After 10 minutes of flight in Film mode where I was testing the ability to stay in a position without drifting away with the winds I changed to Sport mode.
Great wifi signal, 37meters AGL, speed was 12m/s with almost rear wind of 5m/s.
When it was 280m away I saw on the display a tipical free fall case: drone rotating on every axis falling down without control.
GPS position was recorded and the drone found in the bushes along the beach.
Il was emitting an intermittent sound from the motors that was determinant in order to find it.
I donwloaded all the flight log data and watched the video founding something very strange:
- I saw the fall on my phone, from sky to bushes and the video was visible for 2 seconds after it was on the ground BUT the video recorded was stopped at a point when the drone was in the sky and under control.
- the logs showed a partial change in the height record and battery: they looked truncated to 0 just before the fall.
Anafi flew many other times, even with strong winds, and I managed to send it >1km away over the sea to reach some little islands with no issues at all.
In the last week I mounted the modified battery (900mAh 7,6V 60C) and during a drain test (usually 9' 15" until auto landing) I assisted to a fall, inside my house.
Anafi banked left without control and fell, emitting the same intermittent saound from the motors.
The voltage was above 7'500mAh when rebooted.
Then removed the mainboard and double cheched every connector:
- the down facing camera connector has a little way to move on its side and slightly up near the mainboard
- the little flat power connector to the mainboard also could slightly move, even if kept in place by the plastic clip.
I decided to mount some plastic spacers to increase the pressure on those connectors, and did another try to find the problem.
Yesterday I managed another drain test with the light 900mAh battery, and thid time I got exactly what happened:
- during an hovering test i touched every drone arm and found some vibrations, that kind that could easily cause problems on the connectors! (the actual propellers are not the original ones, maybe they are not balanced and it's not an easy task with this split type)
- anafi did a safe autolanding at 9'15" (from 1 meter of height)
- I left the Anafi leaving on the carpet for about 20", with video recording (launched after landing) and the voltage had a very rapid drop down to <5000mV
I was looking the phone, video was still ok during the drop and wifi was working.
Few seconds later the drone entered in the same state I heard after the crashes: intermittent sound from the motors!
I put another battery and downloaded the videos, so the logs.
That's the point:
- when the battery have a drop in voltage WIFI can work and send video from camera BUT the recording stops
- the Logs also are stopped before the drop, and NO CRASH are reported (just like the other two times!).
MY CONCLUSIONS:
If the Anafi fall from the sky, sending video to your phone until down to the ground, BUT the video recorded stops just before the fall, then it's a battery problem related to a current/tension drop. I had proof that Anafi can send video with voltages below 5000mV, wich is not enough to record the crash in the logs or to write in the SD card.
If it falls without sending video to the wifi then the problem could be connector related (power to mainboard due to vibrations).
In the past I had several messages, in FreeFlight, about some 'vibrations'. They appeared just when I used the non original props.
1) Check your batteries for the real capacity and flight time
2) Use only original propellers and check for any kind of vibrations: even if you don't see them on the camera they can compromise the security of your flights!
3) Read the logs, but don't expect to find any data recorded under 6000mV.
Attachments
Last edited:
