Resonance Charactarization

Hardware Multirotors
,
1

I’m currently working on a lightweight quad with large props and thin arms. Because of these thin arms, stiffness is low, and undesired resonances may result.

What I’d like to do is experimentally quantify the resonant characteristics of the frame. To do this I would deliberately excite the system at a certain frequency and observe the amplitude of the response.

As far as I’m aware ArduPilot does not contain this functionality. What’s more, my FC lacks the flash for onboard Lua scripts! I know Mission Planner supports scripting but I don’t expect I’ll be able to make 100Hz commands over MAVLink.

Any guidance on how I might make this happen would be appreciated.

2

Go here and page down to procedure 23 for a novel approach.
7" build Frame Resonance.

3

You could use the “system identification” flight mode, it allows you to select the frequency range to sweep and the amplitude too.

1 Like
4

Thanks a bunch! SYSID is exactly what I was looking for. I’m surprised I haven’t come across that in the documentation before.

5

Using a speaker as a drive source is interesting! Though if I’m understanding it, there’s only one axis of excitation. I’d rather use the onboard motors because that is the source in flight. “Test what you fly, fly what you test…”

1 Like
6

Very interesting topic.
Please share your results as it could be interesting specially for 3d abs frames.
A low stiffness would be a problem for the flight controller, but not necessarily for a resonance issue.
An FEA could give you a clue to estimate the natural frec. of the arm.
You have two main vibration sources. One is an imbalance in the motor, resulting a force in the horizontal plane of the motor. The other one is the propeller pushing air to the arm.