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

Viewing 15 posts - 616 through 630 (of 634 total)
  • Author
    Posts
  • #63836
    jhendrix
    Participant

    I reflashed the firmware and setup a new model on my radio (my previous setup combined lots of operations onto 1 channel, now I split them up).
    All the tuning stuff now seems to work correctly. I can bench tail tune (servoCal) with it disarmed, it will also now arm and I can do the autotune and in air tail tune.
    Issue I have now is the copter seems to be unstable when adding lots of throttle.
    The only thing I can think of now is my servo speed is less than 120 degrees/sec, David’s was 244.

    #63837
    yumemi5k
    Participant

    120 deg/s is certainly an outlier. You might want to check for binding in the tail servo mechanism. Please note that it might be smooth before inserting screw but bind pretty badly after, as the holes might not be perfectly co-linear.

    #63838
    zophon
    Participant

    For the LR, i got a speed of 190dps, and at least another member reported a similar speed. I think that 244dps applies to the baby tricopter (d’aller props, less mass to move, hence greater velocity). I would suggest to go back to the workbench and smooth/sand the tilt mechanism (parts in flat contact and also as previously said, the hole, you should be able to move the screw across both parts of the mechanism almost by hand alone).

    #63839
    PeterPankras
    Participant

    I get around 196 with the Tri LR on 3S with servocal (benchcalibration)

    #64058
    Wally
    Participant

    My post keeps getting deleted for some reason, trying again here:

    About 1 to 2 occurrences per flight, my tri suddenly drifts to the right forcing me to hold a left roll in order to keep it from drifting too far. After a few seconds and some hard turns, it levels back normal. See video link: My Tri-pic

    Before the linked video, I was having bad jello in my HD vid in addition to the drift issue so I thought it’s because my carbon-fiber-stiff-unbalanced props were causing both the jello in the vid and the drifting. I swapped them out for much more balanced props and flew the tri. The video is much better with these props, but the drifting persists.

    Any idea what would be causing this?

    I have a V4 frame, Kakute F4-V2, Aikon AK32 all both from RCExplorer, Sunnysky 2212-13 980KV w/9×4.7 SF props. I followed the usual vids by David, and 6 point calibration. Attached are screenshots of my setup. Unless I screwed up something, I’m sure it has the correct firmware by jihlein, so why does it show no firmware loaded?
    If there’s more information you need to make an assessment or suggestion, let me know.

    I was hoping to incorporate a GPS for position hold/return home as well as resolving the drift issue.
    https://photos.app.goo.gl/ETX1Y9U5pHfhzuff8
    I hope this project lives on, anybody know what the word is with the iNav Devs?
    Thanks to jihlein and everyone else to make this a fun and unique build.

    #64064
    Wally
    Participant

    Sorry, video link is here: Drifting issue

    #64067
    jihlein
    Participant

    I’m going to guess that this drift is in the level mode, and not in the acro mode.

    I think this has become an issue with dRonin and the “newer” ICM20xxx imu chips. They seem to be much more sensitive to noise on the accel side then the “older” MPU6000. dRonin hasn’t added the extra filtering I believe is present in other flight firmwares. I’ve not seen this issue/complaint on MPU6000 based boards, at least that I’m aware of.

    There’s one thing you can experiment with. Under the AttitudeSettings object, there is a setting called AccelCutoff, which defaults to 0.25. You can slowly raise that to see if it improves things. I started with it 0.1 and worked up to 0.25, seeing some improvement, so this might help. I’d guess at increasing it no more than 0.05 at a time. Just be prepared to switch back to acro mode if it does anything funny.

    You have the correct firmware loaded, as evidenced by what is shown on the device. The pane next to it shows what the GCS has loaded for a firmware update. You’re not in firmware update mode so it’s blank.

    It seems doubtful the iNav guys will include the patch I made for the KakuteF4V2 at this point. The updated target file, and the files for including triflight 0p5 and 0p7 are in my git repo for those interested.

    #64068
    PeterPankras
    Participant

    @Wally can you see what angle it shows in the OSD for roll when you try to keep it level with the Horizon.

    I notice my artificial horizon after a few moments after takeoff shows that the Tri is banking ro the Right.. while it is not. Very wierd.. looks like the kind of drift you are describing.
    Since I only fly acro I do not notice anythong in flight.. I still have to try angle mode then, but I never fly anything like that.
    DOes get me worried tough.

    #64069
    Wally
    Participant

    Correct, drift happens in horizon (level?) mode.

    There’s one thing you can experiment with. Under the AttitudeSettings object, there is a setting called AccelCutoff, which defaults to 0.25. You can slowly raise that to see if it improves things. I started with it 0.1 and worked up to 0.25, seeing some improvement, so this might help. I’d guess at increasing it no more than 0.05 at a time. Just be prepared to switch back to acro mode if it does anything funny.

    Thanks, I will give it a try, won’t be til next weekend though.

    Peter: I’ll record through osd to see roll angle next time. My friend next to me who is observing can see it drift and roll to the right. It only happens on my Tri when I’m well within the flight, and not at startup.

    I’ll try the AttitudeSetting adjustment and acro flying to see if there’s a difference.
    Thanks

    #64070
    PeterPankras
    Participant

    @Wally, yes correction I mean Horizon mode not angle mode.
    I always fly acro mode.

    #64071
    jihlein
    Participant

    “I notice my artificial horizon after a few moments after takeoff shows that the Tri is banking ro the Right.. while it is not. Very wierd.. looks like the kind of drift you are describing.
    Since I only fly acro I do not notice anythong in flight.”

    That’s good information, verifies my theory that’s it’s the accels incorrectly biasing the attitude solution.

    I don’t recall, what board are you flying Peter? I’d like to know which imu chip it is.

    In the mean time I’ll take a look to see if we’re setting the imu on chip filters correctly.

    #64072
    PeterPankras
    Participant

    @jhlein I’m using a kakute F4 V2 , the exact setup David uses in the video.
    IMU is an ICM20689 as far as I know.

    #64220
    Wally
    Participant

    Tried adjusting the AccelCutoff with 0.1, but still had the drifting to the right issue, stops drifting when I do some turns and side movements. Then the next day tried with 0.15, same issue. After a few movements, the tri leveled out and this time I switched to acro mode just to see what happens. Hovered ok, then tried to move it forward and slightly bank turn it. Suddenly flipped on roll axis almost upsidedown. immediately switched back to horizon, but it kept tumbling straight downward and crashed. Attached is the aftermath. I must have missed something, just don’t know what.

    1 broken arm and couple broken 3d printed parts, actually not bad for falling nearly 200 feet.

    Only place I can find another woven square tube is Hobbyking so I could replace it, but…..I’m thinking to abandon the Tri, salvage parts and build a 450-500 quad. With the closing of RC Explorer, and unknown future of tri development, I’m out. Been a fun ride while it lasted, thank you for your help, and for this fun project. Happy flying.

    Wally

    Attachments:
    #64242
    PeterPankras
    Participant

    I feel your pain @Wally

    And I totally understand building a quad with so many uncertain factors wit Tri’s now.

    personally I’m a bit sad we never heard anything from David after the release of the Tri LR release.

    #64516
    wynnsquad
    Participant

    Hi there!

    I’ve been away for a long time and am also sad to see RCE is closing down. I can only imagine that the tri business is a tough one though, and I’m glad that I got to be a part of the David Windestal tricopter era.

    I dusted off my V4 and flew it today and seemed to find a solution to my tail-wagging problems I described in #61543 thanks to the replies I received. I ended up setting the following under “Rate Stabilization (Inner-Loop)”:
    Yaw P = 75
    Yaw I = 50
    Yaw D = 0

    I also set ActuatorSettings->MotorInputOutputCurveFit = 0.8 as was suggested.

    The V4 flew beautifully today both LOS and FPV even though it was breezy and I had a lot of fun. Thanks as always for the help and I’m going to continue enjoying my V4 while it lasts.

    Attachments:
Viewing 15 posts - 616 through 630 (of 634 total)
  • The forum ‘Everything about everything else’ is closed to new topics and replies.