I've been working on a tricopter using Arducopter after my successful journey using KK board. Unfortunately there isn't much information on building tricopter using arducopter but it makes it more challenging and interesting. Here is the picture of my tricopter
My copter arms by keeping the rudder to the left and it doesn't change by reversing the rudder which is weird. Worst than that is the movement of the yaw servo. The servo moves perfect using a servo tester but once its connected to the board, it doesn't have full movement. It moves very little and it doesn't change by changing the travel on my transmitter nor changing it in the mission planner.
It kinda feels that the servo doesn't have any movement in the middle (like when we apply alot of expo) however I have 0 expo right now.
When I try flying it, it yawing to right and I have to apply full left rudder to keep it straight. There is no way to get left yaw in this setup.
any comment on this? Or is there anywhere I could get more info regarding tricopters!
Replies
Hi Hadi i saw you're yaw mechanism and i like to tell you that's awesome! Did you bought it somewhere or did you make it yourself?
After some fiddling its as stable as a KK board?? Seriously? Oy, I'm not spending $200 on a board that requires work before being stable as a $15 board!
I have been through the same experience - a homebuilt tricopter that flies nicely with a KK board but behaves like a stuck pig with Arducopter.
My tail servo slams hard one way on power up and being mechanically limited with a big servo, I fear something will break. This seems to be how the servo position is initialised in the software - to zero, rather than midpoint - so reversing in the Tx will have no effect. It requires a software fix.
My servo also has very limited and very jerky travel. I had a very hard time flying at all. Again, in the software, for some reason there is a large deadzone defined for the tail servo. Also, the limits are set quite low. Software fixes again.
It seems that nobody in the current Dev team has a tricopter so the software can't be properly tested.
The options seems to be:
- become a software guru and develop your own fixes
- plead repeatedly on Arducopter until some Dev takes pity and comes to your aid
- build a Quad (which is currently my path ... very reluctantly ... but I want to USE my $250 Ardu-boards).