1. Do not share user accounts! Any account that is shared by another person will be blocked and closed. This means: we will close not only the account that is shared, but also the main account of the user who uses another person's account. We have the ability to detect account sharing, so please do not try to cheat the system. This action will take place on 04/18/2023. Read all forum rules.
    Dismiss Notice
  2. For downloading SimTools plugins you need a Download Package. Get it with virtual coins that you receive for forum activity or Buy Download Package - We have a zero Spam tolerance so read our forum rules first.

    Buy Now a Download Plan!
  3. Do not try to cheat our system and do not post an unnecessary amount of useless posts only to earn credits here. We have a zero spam tolerance policy and this will cause a ban of your user account. Otherwise we wish you a pleasant stay here! Read the forum rules
  4. We have a few rules which you need to read and accept before posting anything here! Following these rules will keep the forum clean and your stay pleasant. Do not follow these rules can lead to permanent exclusion from this website: Read the forum rules.
    Are you a company? Read our company rules

Slow Speed Testing

Discussion in 'DIY Motion Simulator Building Q&A / FAQ' started by JAD, Apr 8, 2020.

  1. JAD

    JAD Active Member

    Joined:
    Jan 17, 2016
    Messages:
    205
    Location:
    Australia
    Balance:
    1,446Coins
    Ratings:
    +141 / 3 / -0
    I started looking at what the slow speed performance was like for my setup.
    I have ballscrew 1605 with brushed motors, Sabertooth, Kangaroo and encoders.

    I noticed it was getting a bit notchy at slow speeds and I could feel this quite easily in the rig in flying sims.
    When just cruising staight and level but just kinda floating around, I can feel this notchiness and its starting to bother me.

    So I got thinking if I ever have a second shot at re building a rig, a performance metric would be how smooth it can be on a slow and smooth test. I think its is an overlooked factor for flying sims.

    Below is my test and Im suggesting a standard of 20mm vertical heave, 15 sec period and no filtering(Im using Mover here).
    upload_2020-4-8_21-40-58.png



    So I ask you guys if you could post a video of your system performing the same test to compare notes.
    Im particularly interested in Odrive's and AMC-AASD15A for my do over in future
    • Agree Agree x 1
  2. JAD

    JAD Active Member

    Joined:
    Jan 17, 2016
    Messages:
    205
    Location:
    Australia
    Balance:
    1,446Coins
    Ratings:
    +141 / 3 / -0
    @Thanos
    Im interested to see how the AMC-AASD15A servo controller looks at slow speed.
    Could I trouble you to see a demo? Pls refer my first post

    @SilentChill @PeterW
    I'd appreciate if we could see how the Odrive performs too.

    Thanks guys!
  3. pmvcda

    pmvcda aka FlyPT

    Joined:
    Nov 3, 2010
    Messages:
    1,846
    Location:
    Portugal
    Balance:
    14,096Coins
    Ratings:
    +2,169 / 16 / -0
    My Motion Simulator:
    6DOF
    @JAD ,

    A video with a test I made some time ago:


    I will make the same test soon
  4. Thanos

    Thanos Building the Future one AC Servo at a time... or 6

    Joined:
    Jul 6, 2017
    Messages:
    1,346
    Occupation:
    Electronics Engineer
    Location:
    United States
    Balance:
    2,692Coins
    Ratings:
    +1,043 / 9 / -0
    My Motion Simulator:
    AC motor, Motion platform, 4DOF, 6DOF
    Here is on FlyPT mover with test parameters are requested:



    Just for comparison, I control the AASD-15A servos with 500 pulses per revolution resolution...

    500 pulses to advance 5mm on 1605 leadscrew (2 revolutions for 10mm)... so for 400mm stroke actuator, for example, it counts 40,000 positions!!! This is 0.01mm step size resolution accuracy.


    And here is a video I made some time ago that shows how smooth the AASD-15A servos run in slow speeds and even 1 value advance by keyboard buttons click on @hexpod software:

    • Like Like x 2
    Last edited: Apr 13, 2020
  5. Thanos

    Thanos Building the Future one AC Servo at a time... or 6

    Joined:
    Jul 6, 2017
    Messages:
    1,346
    Occupation:
    Electronics Engineer
    Location:
    United States
    Balance:
    2,692Coins
    Ratings:
    +1,043 / 9 / -0
    My Motion Simulator:
    AC motor, Motion platform, 4DOF, 6DOF
    Hi Pedro,

    If you can, take a closeup on the motor link like @JAD did, to see visually any stepping present on the rotation of the motor itself along with the same loop settings:

    capture.jpg

    Thanks
    • Like Like x 1
  6. pmvcda

    pmvcda aka FlyPT

    Joined:
    Nov 3, 2010
    Messages:
    1,846
    Location:
    Portugal
    Balance:
    14,096Coins
    Ratings:
    +2,169 / 16 / -0
    My Motion Simulator:
    6DOF
    I will, but you already won :cool:
    • Funny Funny x 1
  7. Thanos

    Thanos Building the Future one AC Servo at a time... or 6

    Joined:
    Jul 6, 2017
    Messages:
    1,346
    Occupation:
    Electronics Engineer
    Location:
    United States
    Balance:
    2,692Coins
    Ratings:
    +1,043 / 9 / -0
    My Motion Simulator:
    AC motor, Motion platform, 4DOF, 6DOF
    Hahaha, Pedro.... Its not pissing contest... lol

    We just need a way to compare each motor solution, and this slow movement test sounds like excellent idea and easy to replicate.

    Thanks
    Thanos
  8. PeterW

    PeterW alias Wickie

    Joined:
    Oct 21, 2018
    Messages:
    199
    Occupation:
    Dipl. Ing. Mb (FH)
    Location:
    Germany
    Balance:
    1,608Coins
    Ratings:
    +360 / 3 / -0
    My Motion Simulator:
    6DOF
    Hi all,
    I just made a quick an dirty test with a motor from my prototype.

    As you can see the movements are not super smooth yet. BUT I never optimised the Odrive parameters in general and for slow speed in special. So too soon to rate finally.
    But so far @Thanos you are the slow speed winner ;)
    • Informative Informative x 2
  9. PeterW

    PeterW alias Wickie

    Joined:
    Oct 21, 2018
    Messages:
    199
    Occupation:
    Dipl. Ing. Mb (FH)
    Location:
    Germany
    Balance:
    1,608Coins
    Ratings:
    +360 / 3 / -0
    My Motion Simulator:
    6DOF
    Hi again @Thanos and @JAD
    This hasn't left me alone and so I have optimized the motor parameters in Odrive a little bit...
    now it runs super smooth :thumbs



    Wickie
    • Like Like x 4
    • Winner Winner x 1
  10. Thanos

    Thanos Building the Future one AC Servo at a time... or 6

    Joined:
    Jul 6, 2017
    Messages:
    1,346
    Occupation:
    Electronics Engineer
    Location:
    United States
    Balance:
    2,692Coins
    Ratings:
    +1,043 / 9 / -0
    My Motion Simulator:
    AC motor, Motion platform, 4DOF, 6DOF
    Good news! Can I ask, how many steps per revolution you get? Or to rephrase, how many data positions?
    I see on the video when its almost stopped, it latches on the closest magnet? When not powered and you rotate the axle by hand, do you feel it latch to these magnetic fields?

    Thanks
    Thanos
  11. Gadget999

    Gadget999 Well-Known Member

    Joined:
    Dec 27, 2015
    Messages:
    1,886
    Location:
    London
    Balance:
    11,543Coins
    Ratings:
    +453 / 9 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, 6DOF
    Is there any backlash in the gears ?

    How many pulses per revolution ?

    It may be a PID issue

    Try a smaller P and I
    Try a larger D
  12. Nisch

    Nisch Member Gold Contributor

    Joined:
    Feb 20, 2019
    Messages:
    49
    Occupation:
    Software Developer
    Location:
    Phoenix, AZ
    Balance:
    266Coins
    Ratings:
    +116 / 0 / -0
    My Motion Simulator:
    6DOF
    I know I'm the odd person out running Teknic ClearPath servos, but here's my setup both connected to the wormdrive and not. My setup is with a AMC1280USB in PWM mode. I can't visibly see cogging, but can barely feel it while in the rig, but I believe that to be the gearbox and not the motor itself.

    You might be able to see it in the reflection on the arm:


    Bare servo:
    • Like Like x 3
    • Informative Informative x 1
    Last edited: Apr 14, 2020
  13. pmvcda

    pmvcda aka FlyPT

    Joined:
    Nov 3, 2010
    Messages:
    1,846
    Location:
    Portugal
    Balance:
    14,096Coins
    Ratings:
    +2,169 / 16 / -0
    My Motion Simulator:
    6DOF
    We should make a table with:

    Slow speed video

    Fast speed video

    Vibration video

    Cost for 6 motors + drives + controllers + power supply for the 6

    This could be a good help for newcomers, to help them decide on hardware.
    • Like Like x 2
  14. JAD

    JAD Active Member

    Joined:
    Jan 17, 2016
    Messages:
    205
    Location:
    Australia
    Balance:
    1,446Coins
    Ratings:
    +141 / 3 / -0
    Looks like we are about to collectively embark on producing a Xsimulator standard! I like it.
    • Like Like x 1
  15. JAD

    JAD Active Member

    Joined:
    Jan 17, 2016
    Messages:
    205
    Location:
    Australia
    Balance:
    1,446Coins
    Ratings:
    +141 / 3 / -0
    Thats good to see. Thanks for posting this.
    • Like Like x 1
  16. Thanos

    Thanos Building the Future one AC Servo at a time... or 6

    Joined:
    Jul 6, 2017
    Messages:
    1,346
    Occupation:
    Electronics Engineer
    Location:
    United States
    Balance:
    2,692Coins
    Ratings:
    +1,043 / 9 / -0
    My Motion Simulator:
    AC motor, Motion platform, 4DOF, 6DOF
    I can make a video that contains all three.
    Any preference on the settings for each on FlyPT?

    Something that we all use to demonstrate with same settings...
    • Useful Useful x 1
    Last edited: Apr 14, 2020
  17. JAD

    JAD Active Member

    Joined:
    Jan 17, 2016
    Messages:
    205
    Location:
    Australia
    Balance:
    1,446Coins
    Ratings:
    +141 / 3 / -0
    Thanks for posting this.
    Very interesting
  18. JAD

    JAD Active Member

    Joined:
    Jan 17, 2016
    Messages:
    205
    Location:
    Australia
    Balance:
    1,446Coins
    Ratings:
    +141 / 3 / -0
    Im hesitant to fool around with the PID numbers. These were all automatically set in the tune. But I will have a shot at it this weekend.
    Its easy enough to restore to a previous tune if it all goes pear shaped.

    In my setup, I have 360 lines per rev encoder, a gear ratio 31(encoder):20(ballscrew) and 1605 ballscrew which results in 46 encoder lines per mm of travel.
    Im using 2^14 bit range (16384) to drive a travel of 355mm, which results in 46 bit range per mm (a fluke that this all aligned neatly)
    So I think the resolution of the feedback encoder and the resolution of the commanded postion is fine enough.

    Another thing I have been playing around with is the dead band on the positioning.
    In my first post video, I have it set to 1 line of position deadband. And this is usually OK but occasionly it would oscillate when sitting there on a contanst hold position.
    Its a very fine oscillation. Its only when Im just sitting in the rig and browsing the web for a while that it gets annoying.
    Sometimes I could just gently dampen the oscillation with my foot/hand, or just shuffle my weight in the rig to make it settle and be completly stationary.
    If I change it to 3 lines of position deadband, then it would be fine and hold completely still at any commanded position.
    But I dont think this small postion deadband is a factor in this slow speed notchy feeling.

    I'll have a play with PID, but I suspect the notchy feeling is a result of brushed permenant magnet motors.
  19. Thanos

    Thanos Building the Future one AC Servo at a time... or 6

    Joined:
    Jul 6, 2017
    Messages:
    1,346
    Occupation:
    Electronics Engineer
    Location:
    United States
    Balance:
    2,692Coins
    Ratings:
    +1,043 / 9 / -0
    My Motion Simulator:
    AC motor, Motion platform, 4DOF, 6DOF
    Vibration test on AASD-15A servomotor.

    Lower period is 50ms to maintain smooth reversal of the servomotor in 1/4 rotation travel. It can go faster of course for lower amplitude



    Here is the mover file I used for testing:

    Attached Files:

  20. Gadget999

    Gadget999 Well-Known Member

    Joined:
    Dec 27, 2015
    Messages:
    1,886
    Location:
    London
    Balance:
    11,543Coins
    Ratings:
    +453 / 9 / -0
    My Motion Simulator:
    2DOF, DC motor, Arduino, 6DOF
    What motors do you use ?

    Try some pid tuning - that may be all the issue is