Dronin on Rcexplorer F3FC integrated board

[jihlein]

Superseded – here’s the latest:

https://1drv.ms/u/s!Am7KDquQANJl5lZoXx6jV0OD10vJ

I just ordered an F3Fc and PDB for my Baby Tri along with the “correct” servo. My servo is slow, and I can’t get 6 volts to it with the existing setup, so……

We’ve had some good reasonable success with the tail oscillations, still not perfect, but I’m looking at a few things. First off is trying to determine if pitch is exciting yaw, or yaw is exciting pitch, i.e. who’s driving the bus? I’ve got a high speed logger I’ll be adding to the Baby Tri when I replace the FC stack.

[MadPoet]

@jihlein what’s your github link, I am on Mac or Linux.

[jihlein]

https://github.com/jihlein/dronin/tree/myDronin

[Kevin_Erik]

@jihlein

The copter will see more tail issues the smaller the frame and higher the thrust-to-weight. I’ve something like 200+ Plasmatree logs that cumulatively show this. In my experience, I’d say Pitch is driving Yaw. Most of the logs show higher step-rate deviations on Pitch that seem to drive Yaw crazy. Did most of my loggin at 4K so if you want, i can upload the lot to G-Drive and post you the link.

[Kevin_Erik]

Here it is, all 800 megs worth Zip’d: https://drive.google.com/file/d/1vchkEmim_LwC5IDgQMng2Ptd7X6Ifa4w/view?usp=sharing

Keep in mind that a lot of this I was experimenting and performing short flights indoors. There are changes to both Filters and PID’s that may not seem to fit but i wanted to try various combinations to get a feel for the process.

Conclusions:

-Pitch seems to drive Yaw as Pitch often has to highest deviation of both.
-Filtering has a huge impact on the response and quality of the Yaw Step response. Too much and its slow and sloppy, too little and your burning up Servo’s. Shitty thing is, that there is no way to tell until the Servo just kinda dies.
-Servo and Yaw filters should be avoided as I’ve seen nothing in the way of performance improvements from either. (Just adds more lag)
-Servo PWM rate and center point should be entered manually if known from the mfr.
-Voltage is critical. 5V is ok but you’ll burn up fewer Servos using 6v and have a much sharper step response.
-Indoor testing to create BF logs, even at 4K sampling is often inconclusive or misleading. Plasmatree logs are best done as a combination of stick “wiggling” (10 actions Roll, Pitch, Yaw, then mixed) and true flight. Otherwise the results will miss alot of what you’ll see in real flight.
-Wind isn’t so much of an issue when taking logs as I’ve not seen any significant differences.
-The BF logs don’t seem to hold any of the Debug data from the Roll axis. None of my logs show this and I know for a fact the Debug “Notch” is active.

[LitterBug]

@madpoet, Yeah, that was from one of my Baby Tris with a BabyPDB and F3FC Racing FC last July. I’ve done a lot more testing, and there have been quite a few developments with dRonin and @jhlein’s fork since then. I had a long period where my DVR wasn’t working, so I don’t have much footage from better flights. Think maybe only some from my mini tri that has a DTFc F3 FC/PDB. I’ve also updated the tail pivot to the Impossible Tilt on all builds. Will be getting some new footage real soon from the current mashup.

I recommend soft mounting the FC on the Baby tri. Made a noticeable improvement on both of my baby tri builds. I run 8mm X 8mm rubber bobbins between the PDB and FC. My VTX/OSD board is hard mounted above my FC, which adds a small amount of mass to the isolated part which also seems to be benificial. I’m running these on both my Baby builds https://www.readymaderc.com/products/details/rmrc-m3-bobbins-8x8mm-black-4pc

[Cshep70]

Litterbug,
Thank You.

[Author]

Posts