Home Forums Everything about everything else dRonin for Tricopters! (TriRonin)

This topic contains 144 replies, has 14 voices, and was last updated by  Kevin_Erik 21 hours, 17 minutes ago.

Viewing 15 posts - 121 through 135 (of 145 total)
  • Author
    Posts
  • #60814

    jihlein
    Participant

    “At this point Iā€™d go so far as to say that the last FCF3 build did the trick.”

    Cool another positive test to add to @krkljan‘s. I think we can call this fixed. By the way, whats an FCF3? šŸ™‚ šŸ™‚

    “I vote to have all sensors working + Crossfire. Nav modes not important.”

    I can do that without affecting Crossfire and Graupner interfaces. In fact I’ve previously tested that very configuration.

    Give me a few days to get it cleaned up and merged. I’ve also got the dRonin PR that integrates DShot ESC telemetry into the F4 boards almost ready, made an initial test on that last night with a BrainRE1 flight controller and Aikon 4 in 1 35 amp 6s ESC and it seems to work.

    I’d like to get both of these merged, and then make a stable release that we can use for a while.

    #60816

    Kevin_Erik
    Participant

    F3FC… From Tricopter Mini & V4

    #60817

    jihlein
    Participant

    “F3FCā€¦ From Tricopter Mini & V4”

    Just giving you a hard time……I can’t type either…….

    #60818

    Kevin_Erik
    Participant

    @jihlein

    Question: Would it be possible to have two versions of the Hex for the “F3FC”. One version thats bare-bones suited for those that “fun” fly and one for those that want all the extras for Navigation?

    Reason I suggest this is that the Servo lag-monster I’m noted earlier could be as you stated, due to the system resource limitations taking effect. Oh, yeah, i did a quick look and my system utilization is 44% with nothing other than battery and MSP Telemetry enabled. Briefly disabling Triflight, lowered this figure by 4% but obviously thats not realistic. Later with the battery finally attached, the total figure was 51% and i can imagine this to get much higher in flight.

    Side Note: Redid the Servo Bench Cal and was able to get the speed Value to “300”. System utilization remained at 51% throughout.

    #60819

    jihlein
    Participant

    Two version would not be my first choice. If the modules aren’t enabled, they don’t have any cost wrt stack use or processor throughput, they just cost the flash space to hold the code.

    The CPU numbers you show should not be an issue. We’ll have to keep an eye on the apparent tail tune lag and see if we can figure out what’s causing it. I don’t think I’ve seen it on the F4 boards, but lets see what happens as the KakuteF4V2 is used more.

    #60820

    Kevin_Erik
    Participant

    Just swapped the protocols around to discover that d-Shot adds 3% to the F3FC CPU utilization.

    #60831

    Kevin_Erik
    Participant

    Completed the 1st flight and Autotune for the F3FC!
    Performed using 4S 1300 mAh Battery and 5″ Dal 5045 Props.

    Stability Tab Pre-Autotune settings were as follows:
    Roll: 15/15/0
    Pitch: 15/15/0
    Yaw: 20/20/0

    Side note: Never once heard the beeper.

    Will Servo Cal soon.

    #60838

    Kevin_Erik
    Participant

    Servo Cal 13.09… Really high but it did Cal.
    I’ve noticed that the DynamicYawHoverPercent either Defaulted or didn’t save. Will have to redo the Servo Calibration at some later point as the weather outside is getting pretty nasty.

    Beyond that, the post Autotune flight was crisp and smooth.

    Attachments:
    #60840

    jihlein
    Participant

    The DynamicYawHoverPercent does not autopopulate, you have to manually transfer that number from the SystemIdent HoverThrottle field for now.

    I probably never explained the change from TailTune to ServoCalibration labels. There were two reasons behind this:

    1)First, it’s not really a “tune” in the way most people think of it. Tune implies PIDs, and this does nothing with the PID or other control loops. It simply sets the fdbk limits (on ground part), and hover trim point (in air part).
    2)Many of the same routines are used for bicopter. The bicopter has no “tail”, just servos.

    So ServoCalibration made more sense in a number of ways.

    #60847

    jihlein
    Participant

    Okay I’ve got the ESC telem integrated for F4 targets, the KakuteF4V2, and Matek405 (and the BrainRE1). It can be used for battery and current monitoring, and appears in log files. I tested it on a quad with the BrainRE1, and it works as it should. At the moment I can’t test the KakuteF4V2 or Matek405, I need to build a bird with the correct components (I wasn’t smart enough to add the telem wires to my BiCopter build, that wasn’t too bright…..). I see no reason why it should not work, as the code is the same among the F4 targets, and the DShot Telem option appears in the hardware objects as it should.

    Next up I’ll get the F3 sensors (baro, gps, mag) added. Then I’ll release another set of install files. Unless someone finds a nasty bug, I think we’ll let things settle for a while with this build.

    #60851

    jihlein
    Participant

    Okay, here it is. F3Fc with the baro/gps/mag sensors, and DShot ESC telemetry on the F4 targets. I loaded this build into all my birds (F3Fc Racing, KakuteF4V2, Matek405, BrainRE1) and bench tested each. All work as expected. While F3Fc Racing behaved correctly, the F3Fc Integrated has a large yaw gyro bias that I can’t null out. It should be noted that this is a crashed/damaged board, so I suspect this is the gyro on this specific board and not systemic to the build. Just be a little cautious and check things out carefully. Weather permitting (not likely from the forecast) I’ll try to fly one or more of these tomorrow.

    There’s a number of new options for the various targets, so make sure all your settings transfer correctly.

    I’m hopeful that this will become our baseline for a while. @david can you post these links in the first post of this thread? I’m going to remove all the older installers in the next few days to avoid future confusion.

    Windows:
    https://1drv.ms/u/s!Am7KDquQANJl5mtP6wXJ6Sdkkm-Q

    MACOS:
    https://1drv.ms/u/s!Am7KDquQANJl5mzvzxSHAVau53J7

    #60853

    Kevin_Erik
    Participant

    Jihlein,

    Any advice on how to use the new installer and hex without having to resetup eveything? In other words, is it possible to save the current copter config and reuse it for the updated hex?

    Idealy, Iā€™d like to avoid having to redo the Tune if possible.

    Kevin

    #60857

    jihlein
    Participant

    With your current version of the GCS/firmware, goto GCS-File-Export UAV Settings and save a backup of your controllers settings, just in case the update goes haywire.

    Now with the new version, plug in the controler. The GCS should recognize that the GCS/Firmware versions don’t match, and ask if you want to update. Say yes, and follow all the steps.

    Now depending on the changes in the firmware, you might be ready to go from this point. But given the changes made for this version, it’s likely the hardware settings will be incorrect. This is because features were added to the hardware configuration, DShot telem for F4 uarts, and the sensor enables for the F3Fc. So carefully look over the hardware pane, make any necessary corrections, save and reboot. Now you should have all your settings, tunes, etc.

    If this get really messed up, go back to your original version, import the the backup settings file you created in the first step, and repeat the process.

    #60861

    Updated the links on the first page šŸ™‚ Thanks for the hard work

    #60878

    krkljan
    Participant

    Something is wrong with GCS. David’s link from the beginning are not working. When using jihlein link and run the file this is what I get:

    Attachments:
Viewing 15 posts - 121 through 135 (of 145 total)

You must be logged in to reply to this topic.