Home Forums Everything about everything else Triflight 0.6 RC1

Viewing 15 posts - 31 through 45 (of 102 total)
  • Author
    Posts
  • #34872
    LitterBug
    Participant

    0.5 and 0.6 status, tasks, and version information from my F3FC

    0.5

    # status
    System Uptime: 3391 seconds, Voltage: 1 * 0.1V (1S battery - NOT PRESENT), System load: 0.09
    CPU Clock=72MHz, GYRO=MPU6000, ACC=MPU6000
    Cycle Time: 992, I2C Errors: 4, registry size: 45
    # tasks
    Task list          max/us  avg/us rate/hz maxload avgload     total/ms
     0 -       SYSTEM       8       0       9    0.5%    0.5%       125
     1 -     GYRO/PID     445     339    1010   45.4%   34.7%   1178028
     2 -        ACCEL     228     187     991   23.0%   19.0%    699805
     3 -       SERIAL    2943       2      99   29.6%    0.5%     33934
     4 -       BEEPER      14       0      99    0.6%    0.5%      1877
     5 -      BATTERY      13       0      48    0.5%    0.5%       899
     6 -           RX      89      58      49    0.9%    0.7%     13675
    # version
    # TriFlight 0.5/F3FC 1.13.0 Jul 24 2016 / 11:56:10 (400
    

    0.6

    # status
    System Uptime: 25 seconds, Voltage: 1 * 0.1V (1S battery - NOT PRESENT), System load: 0.35
    CPU Clock=72MHz, GYRO=MPU6000, ACC=MPU6000
    PIDd: 653, GYROd: 651, I2C Errors: 4, registry size: 51
    # tasks
    Task list          max/us  avg/us rate/hz maxload avgload     total/ms
     0 -       SYSTEM       3       0       9    0.5%    0.5%         0
     1 -         GYRO     250     197    1543   39.0%   30.8%     12726
     2 -          PID     214     160    1543   33.5%   25.1%     10209
     3 -        ACCEL     222     185     799   18.2%   15.2%      6120
     4 -     ATTITUDE      65      33      99    1.1%    0.8%       215
     5 -           RX      62      30      49    0.8%    0.6%       101
     6 -       SERIAL    3351       2      99   33.6%    0.5%       245
     7 -      BATTERY      11       0      47    0.5%    0.5%         9
     8 -       BEEPER       5       0      99    0.5%    0.5%         6
    # version
    # TriFlight 0.6/RCEXPLORERF3 1.14.0 Nov  5 2016 / 17:12:24 (67dd596)
    

    LB

    #34890
    lauka
    Participant

    Okay I made an attempt to fix the CPU load issue from CF. The fix should also fix the problem with the motors getting hot.

    F3FC can now run the PID task at 2 kHz with 33% CPU load.

    I’m throwing a pre-release HEX here as I didn’t do flight testing with this one.

    This HEX also contains:
    – Dynamic yaw was applied twice, removed that. Defaults changed accordingly.
    – In-air tailtune changed to automatically loosen the sampling limits to allow higher success rates on baby tri and other demanding copters. Thanks to Bengt-M for this! If you’re going to test this, please report the min and max values for servo 6 after performing in-air tail tune. We are storing some debug data on those.
    – Default gyro loop frequency increased to 2 kHz.
    – Code cleanup by Bengt-M.

    Please report back if you guys have the courage and time to test this. I don’t have possibilities to test during the weekdays due to winter is coming.

    #34899
    LitterBug
    Participant

    @lauka Awesome. I will give this update a test tonight.

    Cheers!
    LitterBug

    EDIT: PS. Josh said he is trying to work out the tail tune before he posts any Tri tuning videos. 😉

    #34901
    lauka
    Participant

    Nice! Don’t know what kind of issues he’s having, but this release should make the tuning easier.

    Curious to hear how he likes the tri. I bet he gives some feedback about the tumbling when cutting the throttle in FFF :).

    #34906
    LitterBug
    Participant

    I’ll suggest that he poke his head in this thread…. 😉

    Cheers!
    LitterBug

    #34910
    swissfreek
    Participant

    @litterbug Yes when I say stock I mean everything. Exactly the hardware that @David sells on the site, and flying the firmware without changing anything (except for mandatory stuff like calibrating ESCs and accelerometer, the two tail-tunes, and the radio setup, of course).

    @lkaino Sure I’m courageous, and I got out of work early today. I’ll flash it in a few minutes and give it a go.

    #34912
    LitterBug
    Participant

    Just ran with bone stock TF 0.6 RC2 settings. Same deal with the Hot motors. Tried changing thrust factor to 50, Same deal. Now that I know what to watch for, I’m only doing a quick 10 second hover test and not even getting into the tailtune. Beeper isn’t crunchy any more, but the motors and the servo are jumping all around. Getting ready to try “David’s” settings.

    Cheers!
    LitterBug

    #34916
    swissfreek
    Participant

    How does one go about testing CPU load? I have it plugged in and sitting idle, with a battery connected, and it is at ~55% CPU load. Arming it (props off, duh), and then picking it up and moving it around to induce PID responses while still plugged into the computer gets me to ~65% CPU load.

    Putting props back on now to go for a ride. Again, not having any issues with weird sounding motors (yet) or the servo misbehaving. Will get back to you after the flight.

    Oh, and the BlackBox still recorded a blank file. I’ve got it installed and enabled just in case something changes in flight, but not holding my breath.

    #34917
    LitterBug
    Participant

    Ran again with the “recommended” settings, Same thing. Front motors are getting hot, servo is really twitchy.
    Doing Tasks over and over, the Gyro and PID Rate/hz are jumping all over. Still looks like CPU. I’m going to try knocking the Gyro/PID down and see if it gets better.

    Cheers!
    LitterBug

    EDIT

    # version
    # TriFlight 0.6/RCEXPLORERF3 1.14.0 Nov 16 2016 / 18:22:15 (d4d6b2d)
    # status
    System Uptime: 20 seconds, Voltage: 0 * 0.1V (1S battery - NOT PRESENT), System load: 0.32
    CPU Clock=72MHz, GYRO=MPU6000, ACC=MPU6000
    PIDd: 448, GYROd: 449, I2C Errors: 4, registry size: 51
    
    # tasks
    Task list          max/us  avg/us rate/hz maxload avgload     total/ms
     0 -       SYSTEM       3       0       9    0.5%    0.5%         0
     1 -         GYRO     249     198    1709   43.0%   34.3%     10851
     2 -          PID      98      51    1706   17.2%    9.2%      3632
     3 -        ACCEL     219     185     993   22.2%   18.8%      5209
     4 -     ATTITUDE      64      33      92    1.0%    0.8%       156
     5 -           RX      86      55      49    0.9%    0.7%       104
     6 -       SERIAL    3013       2      95   29.1%    0.5%        85
     7 -      BATTERY      12       0      48    0.5%    0.5%         6
     8 -       BEEPER       9       0      95    0.5%    0.5%        15
    
    # tasks
    Task list          max/us  avg/us rate/hz maxload avgload     total/ms
     0 -       SYSTEM       3       0       9    0.5%    0.5%         0
     1 -         GYRO     249     198    1824   45.9%   36.6%     15377
     2 -          PID      98      51    1828   18.4%    9.8%      5145
     3 -        ACCEL     219     185     928   20.8%   17.6%      7388
     4 -     ATTITUDE      64      33      99    1.1%    0.8%       222
     5 -           RX      86      55      48    0.9%    0.7%       147
     6 -       SERIAL   11207       2      93  104.7%    0.5%        99
     7 -      BATTERY      12       0      49    0.5%    0.5%         9
     8 -       BEEPER       9       0      93    0.5%    0.5%        22
    
    # tasks
    Task list          max/us  avg/us rate/hz maxload avgload     total/ms
     0 -       SYSTEM       3       0       9    0.5%    0.5%         0
     1 -         GYRO     249     198    1477   37.2%   29.7%     17349
     2 -          PID      98      51    1479   14.9%    8.0%      5805
     3 -        ACCEL     219     185     902   20.2%   17.1%      8338
     4 -     ATTITUDE      64      33      99    1.1%    0.8%       250
     5 -           RX      86      55      49    0.9%    0.7%       166
     6 -       SERIAL   11207       2      94  105.8%    0.5%       111
     7 -      BATTERY      12       0      48    0.5%    0.5%        10
     8 -       BEEPER       9       0      92    0.5%    0.5%        24
    
    #34919
    LitterBug
    Participant

    Parameters that I have changed from default 0.6 RC2:

    feature -BLACKBOX
    feature RX_SERIAL
    map TAER1234
    serial 1 64 115200 57600 0 115200
    set serialrx_provider = SPEK2048
    set min_throttle = 1050
    set disarm_kill_switch = OFF
    set tri_tail_motor_thrustfactor = 50
    set tri_tail_servo_speed = 305
    set failsafe_procedure = 1
    aux 0 0 1 1450 2050
    aux 1 1 0 1450 1550
    aux 2 2 0 1950 2050
    aux 3 12 2 1450 1550
    aux 4 27 1 1450 2050
    aux 5 28 2 1950 2050
    servo 5 1018 1864 1446 100 -1
    set yaw_deadband = 4
    set p_pitch = 33
    set i_pitch = 22
    set d_pitch = 20
    set p_roll = 39
    set i_roll = 23
    set d_roll = 27
    set d_level = 100
    set rc_expo = 70
    set rc_yaw_expo = 75
    set roll_rate = 45
    set pitch_rate = 45
    set yaw_rate = 80
    
    #34920
    swissfreek
    Participant

    Flies much better for me than RC1. I assume that removing the double-pump on the Dynamic Yaw is what fixed the yaw on punchout, because that doesn’t happen anymore. Tail servo is a little twitchy, but I wasn’t able to run the in-flight tail-tune. I actually got the fail tones (that’s a first for me). When I got it back to the bench I noticed that the motor thrustfactor was still set to 13.8. So I set it to 5 and will now go reattempt the tail tune.

    No issues with hot motors or anything like that.

    #34921
    LitterBug
    Participant

    For Shiggles, I put some lower pitch props on. Heat is gone. Going to bump Gyro/PID back up. Was running fine with these same props every time I went back to 0.5. Maybe something in the filtering

    Cheers!
    LitterBug

    #34923
    swissfreek
    Participant

    Tried to do the inflight tail-tune again, this time with tri_tail_motor_thrustfactor set to 50. Again I got the double-beep failure. Setting it to five got rid of the twitch in the tail, though, and it flew pretty stable and similarly to what I was used to on 0.5, so I’m guessing that 5 is once again in the ballpark like it used to be.

    I double-checked my bench tail-tune and confirmed that the servo center and end points are as close to perfect as I can get them (visual level to top plate for center, used the little tool @David sends for left and right). So I’m not sure what exactly is going on with the in-flight tail tune.

    Running out of daylight now, so further testing and maybe some FPV even if I can’t get the in-flight tail tune to work will have to wait until the weekend, most likely.

    #34924
    swissfreek
    Participant

    Oh, also, all my servos read the default values (1500/1000/2000) except for Servo 5. That reads 1574/1166/1975 (I assume that’s the result of the bench tail-tune). Of course I didn’t successfully complete inflight tail tune so I’m not sure what you’d expect to see there.

    Also, BlackBox did log a file during flight, with data in it, but it’s gibberish and isn’t understood by the CleanFlight BlackBox viewer.

    #34925
    swissfreek
    Participant

    This is what the BlackBox file looked like…

    Attachments:
    1. LOG00064.txt
Viewing 15 posts - 31 through 45 (of 102 total)
  • The forum ‘Everything about everything else’ is closed to new topics and replies.