Yes, I should have qualified my statement about aluminum. A monocoque or shell structure would work. Like your printed frames. Stamped though.
We did some Dev work with printed 17-4PH stainless, again unrelated to the hobby, with success. But, crazy expensive.
“Stamped though” I got to see if i can rent a mega stamp.
Heh he. Seriously, a Hass could do it. I made a lot of MDF frames on my CNC.
Made of MDF, if any one would like to mill this frame let me know.
Log: https://drive.google.com/file/d/1Xh_i-ZhgHGPLpn1-vqnQFFdQ7vQArvv5/view?usp=sharing
Finally, after a period if bad weather. I finally got to fly the version with baffles and solid arms. The stiffest one yet.
Noise in the RATE controller seems well managed. Keep in mind there’s no filters applied here either.
The vibrations loos good sitting at just under 5 at hover but the Y axis spiked as high as 16.8 during a punch/fast ascend. I’ll try the one with added support beams next just to see if there’s any difference. The spike is still a bit too high for my liking.
The yaw still deviates at some points but since it’s not tuned and the GPS/compass unit sits right o top of my ESC it would be some interference ruining my tests on that one. Not as bad or consistent as before so the better torsion resistance might have helped out a bit.
As for the X and Y vibration spikes. I guess it might be wires. The frame with support beams limit the arms X and Y movements so it might help as well.
Here’s a pic of the Fc mounting. The GPS and ESC wire enters from the side, might have something to do with the Y axis spiking the highest.
board: Matek H743 smil V3
I have lower vibes on the secondary IMU. I just wanted to ask if different update rates could affect the peaks of the vibration detected. The 1st IMU has 2000 Ghz and the second has 1000 under the IMU tab in log analysis.
And, would I be better off using the second. Second has around 2.5-3 at hover and first has 5-6 around hover. The second IMU wont peak at movements like the first. Is that also du to a lower measurement rate? I won’t just not peak as high but sometimes not peak at all. That I suppose is a resonance issue of the first IMU though.
It’s helpful to see the whole picture of you post a .bin file
Because fast rate is only enabled on the 1st IMU.
I don’t think so. It’s probably just not catching some dynamic peaks from the lower sample rate.
Ah ok, so it’s my logging settings and not the IMU itself.
Good to know it’s mechanical, so I don’t go in the wrong direction trying to solve it.
Thanks!
Not logging, this parameter INS_FAST_SAMPLE,1
I suppose this is default for that Flight Controller unless you changed it. It’s interesting because the Matek H743-Slim I had (V1.5 I think) had both IMU’s at fast.
I wonder if the difference in the Vibe data from each IMU is an artifact of the Vibe algorithm with respect to the sample rates. Because I looked at the IMU Accel data from each IMU in one of your previous logs shared on this post and they overlay practically identically. Although, that might just be a logging sample rate issue.
I could try setting fast sample on both IMUs.
Do you think I would benefit more from a faster sample rate or lower vibrations and probably noise as well than?
