I just finished building my first little Tricopter (RCExplorer Baby Tri clone 5") and just wanted to make sure the configuration is more or less correct. Actually I’m quite sure it probably isn’t yet, but the Wiki page (Tricopter Configuration — Copter documentation) doesn’t say much about the differences between quads and tris. For example, is Autotune supposed to work just like on a quad?
Here is a logfile of a first attempt outside, which I aborted when I noticed incorrect AHRS orientation (it’s always the basic things you overlook…). Maybe it’s still useful to those who know, so my chances of success on a second attempt tomorrow are maximized: WeTransfer - Send Large Files & Share Photos Online - Up to 2GB Free
Are there any other obvious mistakes in configuration?
Thanks!
On a side note, I noticed that the copter can arm (via RC) on the bench even with zero sats. This happened with two different custom builds, I guess it shouldn’t be possible at all?
Also I’m getting no ESC telemetry in the OSD, although the data is visible in MP. It’s my first time using BDshot with AP, but I assume the OSD elements should appear just like with UART ESC telemetry?
On my next aborted takeoff attempt, the copter wanted to rotate clockwise, so I assumed the tail servo might need to be reversed, to counter that movement correctly.
But now the craft is rotating counter-clockwise on takeoff. Also it looks like the tail servo is exceeding MOT_YAW_SV_ANGLE which is set to the default of 30, slowly leaning further and further to the right, until the tail motor prop scratches the ground. Motor directions are correct (all CCW) and so are the props.
Any ideas, anyone…? I’m sure it’s just a matter of configuration, but as said before, the wiki page on tri’s doesn’t give any more details.
Unfortunately I can only test it inside for now due to bad weather, so I didn’t feel like risking it spiralling out of control in a small room.
EDIT: Ok, reading the wiki page again, I found what I had overlooked:
SERVO7_TRIM: tail servo’s PWM value close to what is required to keep the tail from spinning.
I had assumed the copter would determine this automatically - but in order to monitor rotation, a compass would be mandatory I guess, and that’s not mentioned. (I have none installed either.)
So I guess the correct way would be: To observe the copter, adjust yaw to keep it from spinning, then note the value in channel monitor on the transmitter and enter it as trim value?
To avoid ground effect influence it should probably happen at reasonable altitude, so I guess I won’t try that indoors.
This was assembled from mostly leftovers, including an old GPS without compass (but locks on surprisingly well). I doubt this has to do with missing compass, from what I read one of the reasons might be too high yaw gains (?).
On plane, too high gains on a servo can easily be adjusted simply by launching Autotune, but I guess that’s not an option in this case? Nevertheless, tuning the tail servo might be part of Autotune on a tricopter?
I’m beginning to realize Tricopters are seriously out of fashion - almost every search result on the web is 8-10 years old.
Well, too bad this wasn’t mentioned on the tricopter wiki page - F405-CTR has its I2C pins below the board, so this means some disassembly work. Mandatory compass is also only mentioned for VTOL tailsitters on the compass-less wiki page: Compass-less Operation — Copter documentation
I tried compass-less on a normal quad and it worked ok, so I thought I’d simply use that leftover GPS here. This copter was mainly intended for LOS flying anyway.
