Home Forums Everything about everything else Triflight 0.7 Beta 1

This topic contains 142 replies, has 32 voices, and was last updated by  bhuism 2 years, 7 months ago.

Viewing 15 posts - 1 through 15 (of 143 total)
  • Author
  • #37145


    Triflight 0.7 Beta 1 is available at: https://github.com/lkaino/Triflight/releases/0.7-beta1

    Please post any feedback, comments, discussion related to the release to this thread.

    I have tested the release only on RCEXPLORERF3 and BETAFLIGHTF3 targets, feedback from other targets is welcome.



    SWEET! Thank you for all your efforts @lauka!




    Pids for big tri?



    Great work Lauka! I’ve tested flashing on the RCExplorer integrated F3 board which works fine (just hovered it in the living room)

    I’ve also flashed the following naked FC’s successfully

    Flip32+ (NAZE target)

    On the REVOLT target, the default setup is a QUAD1234 not Tricopter?



    Tested on CC3D and the servo pin is not mapped default.



    Just flew with moderate wind and the origional (first motors etc) mini-tricopter.
    Significant yaw to the right when I lower the throttle during flight including easy cruising around.
    Did a full setup incl tailtune



    @EA: Could you try setting tri_motor_acc_yaw_correction to 0 in CLI and see if it affects it. It sounds like it’s overcorrecting for the motor spin down torque. Do you have active braking enabled in your ESCs?



    Hi Lauka,
    At the field now



    Hi Lauka,
    At the field now, set the tri_motor_acc_yaw_correction to zero. No more yaw to the right now but quite a lot to the left when I cut the throttle after a punch. However much more manageable during cruise.
    I think my activebreaking is on.



    The left yawing is what the parameter tries to fight against. Try to find the sweet spot for your copter.



    Hi Lauka, thanks for that last update you made available to include some F4 boards, much appreciated.
    I have a quick question for pretty much anyone familiar with tricopter setups.
    My question is. I seem to be able to setup Tri-Copters with Naze32 Rev 5 boards running cleanflight and get a pretty ok flight out of it.Seem the flight controller use by default Pins 1 or 2 for servo and 345 for motors,
    But this far i am not able to setup any tri-copters successfully using other flight controllers.
    I have tried KISSFC, F3 SPRAcing, Xracer F303, and lastly AirbotF4
    All boards except the KISSFC was running betaflight
    It seems like I can get 3 motors to spin but never can i get the servo to work. Its like its not receiving a signal.
    Thus far I am not sure if its wiring issue or a software issue
    Do i have to manually set the channels via CLI or is there some other magic i have to do.
    In cleanFlight it seems to all just work, but i only have the Airbot FC left over and it does not have CleanFlight firmware to test with

    Any help is much appreciated.


    Dave Mak

    While I’m not lauka, I might be able to help point out a possible solution / area to look into.

    If you’re using the betaflight 3.1.x releases, you need to set and check your resource mappings.

    I just got a FliteTest Electrohub tricopter I had originally built with a MultiWii Flip 1.5 FC updated to use a SPF3 FC and had this resource mapping work for me:

    resource MOTOR 1 A06
    resource MOTOR 2 A07
    resource MOTOR 3 A11
    resource SERVO 1 A12

    Just note that I have my servo connected to where motor 4 usually is output for a quad… also note that the pins referenced will vary based on the board you’re using.

    It flies great, but I’m still tuning it. I started with PIDs of 40,10,10 on roll and pitch and am working up to something I like. I ran out of time before I had to stop tuning with both roll and pitch P at about 80.



    Thanks Dave 🙂 I was hoping anyone would reply as this is my first look at tri-copters and figured perhaps mapping of some sort needs to be set…but did not know how to go about it.
    I will definitely look into what you suggested.
    The firmware version I think is based on Betaflight 3.1.6, as its the Triflight 0.7 just uploaded by Lauka and i think it was mentioned as a fork of 3.1.6

    Thanks again for the help/advice and I will reply back with my result on this.



    Thanks again for the help Dave, but unfortunately it does not seem to play nice.
    When i have mine as this i lose motor 3 for some reason. I played around with enless combinations of settings within the resources list but no go. Either i lose motor 3 or 2 and 3 when arming and doing a spin up test
    Never do i really get the servo to react when i move the tri-copter from side to side.
    Verified power is good from the output pins, and servo is ok too. Got me stumped, going to give this a break for a bit cause i am not getting anywhere 🙂

    # resources
    resource MOTOR 1 B00
    resource MOTOR 2 B01
    resource MOTOR 3 A03
    resource SERVO 1 A02


    Dave Mak

    PieterC, I would recommend looking at the code specific for your flight controller, so for a naze32:


    And using the resource mapping documentation:


    If I were to connect my motors to the 1-4 output pins, with 4 being the servo, I would setup like such based on this code:

    DEF_TIM(TIM1, CH1, PA8,  TIM_USE_MOTOR, 1), // PWM9 - OUT1
    DEF_TIM(TIM1, CH4, PA11, TIM_USE_MOTOR, 1), // PWM10 - OUT2
    DEF_TIM(TIM4, CH1, PB6,  TIM_USE_MOTOR, 1), // PWM11 - OUT3
    DEF_TIM(TIM4, CH2, PB7,  TIM_USE_MOTOR, 1), // PWM12 - OUT4
    resource MOTOR 1 A08
    resource MOTOR 2 A11
    resource MOTOR 3 B06
    resource SERVO 1 B07

    You can see how I arrived at my mapping for the SPF3 based on this code:

    DEF_TIM(TIM16, CH1, PA6,  TIM_USE_MOTOR, 1), // PWM1 - PA6  - TIM3_CH1, TIM8_BKIN, TIM1_BKIN, *TIM16_CH1
    DEF_TIM(TIM17, CH1, PA7,  TIM_USE_MOTOR, 1), // PWM2 - PA7  - TIM3_CH2, *TIM17_CH1, TIM1_CH1N, TIM8_CH1
    DEF_TIM(TIM4,  CH1, PA11, TIM_USE_MOTOR, 1), // PWM3 - PA11
    DEF_TIM(TIM4,  CH2, PA12, TIM_USE_MOTOR, 1), // PWM4 - PA12
Viewing 15 posts - 1 through 15 (of 143 total)

You must be logged in to reply to this topic.