Hi everyone,
Our team has recently been working on a gas-engine pusher-type VTOL UAV and we’ve been running flight tests. While the aircraft performs well during horizontal flight, we encountered an issue during the transition phase. Specifically, when transitioning from multicopter mode to fixed-wing mode in Auto mode, the aircraft suddenly drops altitude—by nearly 50 meters. (90m to 42m)
For reference, the pitch angle dropped from about 1° to as much as -20° during this event.
red: altitude
green: pitch
This behavior occurred exactly at the moment when the transition completed and the lift motors were shut off. Thankfully, the aircraft recovered and climbed back to the commanded altitude, but for a moment, we genuinely thought it was crashing.
I came across a blog post where someone mentioned this might be a firmware bug. They said upgrading to ArduPlane 4.4.x resolved the issue. However, since we’re already using 4.4.4, I don’t think that explanation applies in our case. (refer to the link: QuadPlane VTOL loses altitude during transition from VTOL to Fixed Wing)
The reason I’m posting this under the 4.6 topics, even though we’re using 4.4.4, is because I’m curious if anyone who’s experienced something similar was able to resolve the issue after updating to the latest firmware.
If this turns out not to be a software issue, we’re planning to try shifting the CG further back and setting Q_TRANS_PIT_MAX to 0 to see if that mitigates the problem.
can you provide the bin file?
Sorry for the late reply. It took some time because I didn’t have the log on hand.
You can download it from the link below.
Hi Yonguk Kim,
From analyzing the .bin log file, it seems that when switching to fixed-wing mode, your aircraft’s CG is far forward. This causes a nose-down tendency right after the transition.
As shown in the image below, the elevator consistently stays at a high deflection to maintain level flight.
1 Like
Thank you very much for your reply.
For your reference, our aircraft has a Mean Aerodynamic Chord (M.A.C) of 334 mm, and we flew with the C.G. set approximately 24.5% from the leading edge. Following your suggestion, we moved the C.G. back to about 27.5%, and this time, the altitude drop was limited to only around 20 meters. We plan to test again with the C.G. set at 28.5% to observe the overall trend.
I have one question. Currently, my Q_TRANSITION_MS is set to 5000 ms (5 seconds). If I increase this value to 10 or even 15 seconds, would it help reduce the altitude drop from 20 meters to around 10 meters during the transition? My assumption is that during the 5-second transition period, the aircraft may not have been generating enough thrust and lift to support its own weight, which could have caused the sudden drop as soon as the motors shut off.
You can download the log(the current flight) from the link below.
I didn’t see the notification, so I’ve just now seen your post.
Have you tried with the CG at 28.5% yet? From what I can see, the elevator is still being held at a certain level (though less than before) to maintain level flight.
As for Q_TRANSITION_MS, in this case it’s not really necessary, since you’ve already set:
ARSPD_FBW_MIN = 24.000000TRIM_ARSPD_CM = 2500.000000
Your aircraft has already reached the required speed of 24 m/s, which is sufficient to trigger the end of the transition.
EDIT: I think you should accurately calculate the CG position of your vtol
