Forum Replies Created

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • in reply to: BiCopter Tuning – BetaFlight #61593
    dmyers7
    Participant

    If you’ll check my previous posts you’ll see that I was onboard with the Bicopter from the beginning last year and helped to route out original mapping for the kakute FC. I applaud everyone’s efforts to improve the bicopter with all kinds of sOAT mods and ducts but I’m intested in getting back to tweaking the stock setup whether hardware placement or software tuning. I’ve been following these threads, but have not seen this yet. The video linked above is not an ad or someone promoting a competing product and he even references David’s site. What I found interesting is he seams to have a relatively good tune on a pseudo rcexplorer design but his FC is hanging upside down and below the carbon arm. Upside down is no big deal, but how far away the FC is from the center of rotation is intriguing.

    So no, I’m not promoting anything but trying to improve the stock setup with minimal changes to the baseline design which hopefully feeds back a result that Dave can release a V2 on. Think of it as a new github development branch.

    Thanks for asking.

    dmyers7
    Participant

    Thank you. Will give this a try hopefully tomorrow. Noreaster coming through, weather has not been a friend.

    dmyers7
    Participant

    Flew it today with it being pretty good wind on both the HQ props Dave sells and a pair of DAL Cyclones 5046 (much more rugged; especially when you are smacking the ground playing around). Both did great and flew really smooth! I really like the Kakute flight controls with the foam dampened accelerator / gyro and fly them on my quad with similar smooth results and longer battery life. It does seem to help isolate from prop and motor vibrations and reduces over reaction and heating the motors. I’ve flown with some really banged up props and still it flies smooth.

    For some reason my black box did not record across all my flights so will have to figure that out and try again; so nothing to share there.

    Had the same basic results as Dave with what seams like reverse acting up. In forward or even sideways (aileron) flight it is smooth once you get moving. I was doing a high speed left banking turn around me and kept increasing throttle (speed) and tightening the circle and it was absolutely rock solid! The PIDs David provided worked really well until you get into one of these prop wash or funky modes; definitely fun to fly!!

    As I changed PIDs (before finally going back to the stock Dave PIDs) I noticed the expected behavior of bad PIDs. Another observation and something that leads me to thinking we may need a complete different control model is when I tried to fly it in acro mode, pushing forward translated the copter motors to tilt forward followed by the body swinging beneath the rotors until the rotors basically face up again, but of course the weight is now forward again so the weight of the copter (the body) swings back. The servos stay tilted because it is acro mode, so you can stop the swing if you add throttle just right. Consider that 2 of the 3 movements rely on servo movement (Elevator & Rudder) and not motor speed, unlike a quad that depends on all motor speed, and the Elevator is dealing with thrust moving forward and back of the CG, effectively a shifting mass. This is very different than a quad where the flight model has the lines of trust on the outside perimeters of mass with the mass central; the mass does not move with reference to where the thrust is produced. Apart from balancing the thrust to keep movement smoothed and locked in, the flight model uses pairs of motors to translate thrust to movement around an axis. When the bi-copter tries this simple model it needs to deal with shifting weight forward and back as the servos move the trust forward and back around a central CG. This is mainly in Elevator movements. Rudder is equal forward and back and Aileron is more like a quad and does not display the same shaking. I noticed that if you get shaking you can power out of it which I think is similar to the normal stick movements when we push forward and add throttle. If you just push forward from a hoover it will shake.

    I bet if you could shift the battery with the servos tilting the motors it would work almost flawless.

    Otherwise a great model, a great start, and fun to fly. Can’t wait to see what the other teams are doing and may noodle this and try some custom code myself.

    dmyers7
    Participant

    I was wondering about that one. It flies by so fast and I never go back to look. I’ll update my post and pull that out.

    dmyers7
    Participant

    That was the issue. Use this mapping instead of Dave’s until he updates it: (Matches the Wiring diagram he provided)

    resource motor 1 B00
    resource motor 2 B01
    resource motor 3 none
    resource motor 4 none
    resource servo 1 A03
    resource led_strip none
    resource servo 2 C08

    Late and Raining or I would give it a try. Did hop it around in the back patio and all controls are in the correct direction! Can’t wait to give it a try. With BetaFlight as the FC the PIDs can be tuned from my Frsky radio so I can tweak until it feels good. Flying with the Orange HQs first. Looking at David’s video, looks like maybe what we need until we can get fancy with the feedback is a PID term reducer when you put in reverse elevator to desensitize the reaction. With the Black Box on the Kakute we should be able to record and see what’s going on. An adventure ahead.

    • This reply was modified 5 years, 4 months ago by dmyers7.
    • This reply was modified 5 years, 4 months ago by dmyers7.
    dmyers7
    Participant

    I have the Kakute F3 V2 board also and can’t get my RCExplorer servos even moving.

    I used the mapping and full PID settings provided by Dave using BetaFlight 3.5.2.

    Here is what I’ve noticed:

    Motors are working
    Dave’s setup does not use BiCopter, it uses “Custom Airplane”
    Using BetaFlight 3.4.1 is not an option, the Accel / Gyro jitters and drifts (is fine in 3.5.2)
    I put the servos into live mode and they don’t come alive, I can still move them manually.
    6V & ground is present. Feedback and Signal is per the pinouts on this site; no not mixed up.
    I will pull out the O-Scope next to look to see if the PWM signal is present to the servos.

    Per Dave:
    resource motor 1 B00
    resource motor 2 B01
    resource motor 3 none
    resource motor 4 none
    resource motor LED A02
    resource servo 1 A03
    resource servo 1 C04

    Seams like this could be incorrect since “servo 1” is mentioned twice.

    Plus what jhlein says it should probably be servo 2 C08

    Where can I get, find, mine the port mapping for the Kakute F4 V2 board? so I can maybe fix / do my own mapping?

    I can now see that since the servo 1 was mapped twice above this may be the reason mine is not doing anything; since C04 is not used in our setup for servos.

    Any help would be appreciated. If I figure it out, will post my findings.

    • This reply was modified 5 years, 4 months ago by dmyers7.
Viewing 6 posts - 1 through 6 (of 6 total)