Hi,
I am not getting the tune to work on a large, 90kg quadcopter. Linked is the bin file from one test. I managed to keep it hovering about 1m above ground with a lot of control input. I have lowered the PIDs a lot but it is still struggling.
It is a large 3m diameter X-quad with 51" props, KV44. 4 x 24S batteries are strapped to the arms so there is a lot of weight far from the center of gravity. Arducopter v4.4.4
Any suggestions on how to change the PIDS?
Use "ArduPilot methodic configurator software " to guide you through a proven sequence of steps guaranteed to get you a working vehicle
Thanks. I just did. It gave me higher PIDs than before. I will try and see how it works out.
Post a .zip file with the files generated from the ArduPilot Methodic Configurator"
The new setup using your tool dramatically improved the stability. Attached is the resulting param-file.
We still experience a slow “vibration” in pitch and roll with about 0.5Hz. Makes me wonder whether to decrease the PIDs or modify any other parameter, see picture from MissionPlanner analysis.
Paramlist 2024-07-24_afterFT.param (19.8 KB)
We need the .zip file and a .bin log file.
Bin-file here. 2024-07-24 14-52-36.bin
Zip-file to follow shortly.
I strongly suggest you fix the battery monitor → voltage must be wrong and current is basically not working at all.
Really you should just get a real voltage and current monitor to suit your copter before doing more flights. Tuning will be incomplete without it, and safety is compromised.
Vibrations are high and you’ll need to fix the physical issues that are causing that in order to end up with a reliable tune.
These will be essential
ARMING_CHECK,1
FENCE_ENABLE,1
LOG_BITMASK,180222
After you’ve fixed the battery voltage monitor, go back to the Initial Parameters calculator and enter your battery cells and prop size, select “Suggested settings” and accept all changes offered.
Any progress on this?
You do not need to do this if you use the configurator component editor window correctly. It does it for you automatically.
I can only see if this was done correctly after the user provides us with a .zip file with up-to-date “reason changed” fields.
Thanks for your support!
The battery voltage monitor is only monitoring the 4S battery powering the cube with peripherals. Each motor has its own 24S battery pack, not monitored in the current set-up. It will be later on.
As we are only doing short flights, using 20-30% of the battery capacity, we have not prioritized this.
We also noticed that the vibrations are high primarily in the Z-direction. We will try to cure this with dampers.
Here is the ZIP-file.
Quadcopter
I would be very careful with that set-up, it can lead to unbalanced discharge (pack to pack not cell to cell) negatively affecting controls especially on low battery and at high thrust demand. IMHO it would be better to have packs connected in parallel through at least ideal diodes, preferably a dedicated controller.
Before throwing dampers at the problem I would look for any mechanical issues that may be causing high vibrations. Since Z is dominant you may have nailed an antinode(s) of the base plate but it may be a serious issue.
If you are using an internally isolated FC like cube adding external vibration isolation may cause increase in vibrations if you get stiffness wrong or are unlucky. External vibration isolation also exposes FC to vibrations transferring through connecting wires.
That will mess up the PID gain voltage scaling algorithm. So not a good idea. You should monitor at least one 24S motor battery on your primary battery monitor and leave the 4S flight controller battery for a secondary battery monitor.