I've been struggling to diagnose a problem with my drone losing altitude in multiple situations.
The first situation where it loses altitude is in auto mode, when taking corners, the drone stops, and then drops up to 6 or so meters, before turning around and regaining the height. This doesn't always happen, and it isn't always this severe when it does.
I've read on this forum and elsewhere that this if often the result of pressure changes on the barometer due to the sudden stopping. Causing the drone to briefly overestimate its height and attempt to adjust by dropping. I tried multiple things to solve this problem like removing the cover from the body, switching MP's EKF_ALT_NSE to rely more heavily upon the GPS, or switching the primary altitude sensor to the GPS. Removing the cover seemed to help quite a bit, but relying more heavily on the GPS did not seem to help.
The second instance where altitude is being lost is when entering loiter/stabilize after flying for a while. The voltage *seems* to be OK at these times, but the drone slowly drifts into the ground, bounces up (often not as high) and then drifts back down. Manual and GPS modes also experience this problem after the drone has been in the air for a longer while. Again, voltage readings are still good for our motors at these times, so I'm not certain what is going on.
Here is the flight log from a debug run I ran today. The weather was fair and wind mild. I flew an auto mission for the first 7 minutes without the cover on during which the dropping was minimal except towards the end. After that I flew several shorter missions with parameters being adjusted during each one. During these shorter missions the dipping became gradually more pronounced. Is it possible that the battery is somehow not pumping out enough juice? Please send help!
A note on my build: Pixhawk2.1 hexcopter with a cheap GPS and DJI E800 propulsion system on 1 4S battery. (Do I need a 5S? 6S?)
After running many more tests today, I have determined that the battery is not the issue. When undertaking aggressive turning maneuvers, or simply sharply turning in general, no matter the battery voltage, the drone will drop. This happens in all modes. The height loss was particularly prevalent when performing an in-place spin, after which the drone would ALWAYS fall.
Furthermore, I'm *fairly* confident that the issue is not with the barometer as adjusting all ALT settings involving the barometer have no affect.
What could this be then? I'm beginning to think that the accelerometer or gyro is possibly fried or out of whack, but this will take more testing to confirm. If anyone has any insight, please share it with me!
After many more tests, I have determined that it is *somehow* a problem with the battery voltage (or how the Pixhawk responds to it). After a LOT of tests, it is always the case that the dipping gets worse the longer the drone has been in the air (AKA lower voltage). I thought that my system was good for these lower voltages, but I guess not? I guess the next thing to try would be a higher voltage battery?
If someone can confirm that I am just an idiot and using too low of voltage battery, that would be great.
Or if anyone here deigned to say anything helpful ever, that would be great too.