servo madness - Page 2


Page 2 of 3 FirstFirst 123 LastLast
Results 21 to 40 of 50

Thread: servo madness

  1. #21
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch1-2-png
    current limiter pot increase



  2. #22
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch2-1-jpg
    servo madness-ch3-1-jpg



  3. #23
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    I am going to wait for some help before I continue, these are the current moves.
    I need to know if I have to work with the servo drivers or work with the software setup



  4. #24
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch3-2-jpg
    so minus gain on input flipped the motion direction



  5. #25
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4043
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    Hi jensenjim,

    Ok great it seems you are now responding to my posts, describing what you are doing, what you see, and asking intelligent questions..

    Moving the carrage in one direction moves the encoders in the same direction
    ok now we know both encoders are counting in the correct direction.

    On a side note in earlier posts you had the system operating somewhat properly. But now the parameters are set to basically crazy nonsense values (no PID and huge Feed Forwards). But that is ok for the moment. Before changing anything I would like to first understand and fix why there is basically opposite behavior between the two axes. Axis #1 moves the correct direction:

    servo madness-right-png


    But Axis #0 moves in the wrong direction:

    servo madness-wrong-png

    Can you see this?


    i can see the ential movement direction and ch 0 has none.
    I don't understand why you think Axis #0 has none. Rather it moves the wrong way and in fact much more - about - 800 counts rather than +100 counts like Axis #1.



    do I still need to work with the servo drivers or can this be corrected with the kmotion
    Most things can be corrected within KMotion. But your amplifiers seem to set very different because the same ~1V (200 count) pulse into each causes Axis #1 to move ~800% more. You might carefully re-check all the drive settings and pots to be sure they are set similarly as it appears you have something set completely different.


    The plots above are the results from a +200 DAC count (exactly the same) pulse to both both motor Drivers. There are many ways to reverse the response of a system such as mechanical linkages, electrically, or in software. I don't know the reason they are different on your system but we should be able to make them the same by changing the OutputGain of Axis #0 from +1 to -1. Please make that change and try the experiment again.

    I'm also guessing you have different IIR Filter settings for the two axes. You didn't show or tell us what they are so we don't know. I see some minor differences in the output (green plots) which should be exactly the same as your feedback terms (PID) are all Zero. Please post your IIR settings for both axes. Verify that you did indeed have them set differently, then clear all the filters for both.


    the two moves neet to be the same before coupling
    I don't really understand what you mean, but yes both axes must be functioning correctly before coupling them together mechanically.

    q1 what determines the blue line
    The blue line is the commanded trajectory. It represents what we are commanding the axis to do. In this case we are commanding the axis to move 300 counts at the specified Velocity, Acceleration, and Jerk

    q2 is the green line the actual voltage
    Basically yes. But it is in units of DAC counts where 2047 counts = 10V. So in your plots the green pulses of 200 DAC counts (use right side scale) are applying approximately 1V to the Drives.

    q3 the move command is a preset command that moves the servo fwd then back
    Well yes, but it is not preset. It will be of the size you specify and with the Velocity, Acceleration, and Jerk you specify.

    why is the blue line different from ch0 and ch1 , I disable the servo, reset the zero, and then enable the servo before I hit move
    They are actually exactly the same (+300 counts on left scale). But the plot auto scales in order to completely show the red and blue plots. Because in one case the axis moves the wrong way the plot zooms out so you can see it.


    the servo driver has pots for
    1 loop gain
    2 current limit
    3 ref in gain
    4 test/offset
    I'm not that familiar with your Drive. Eventually you will want the Ref in Gain to be set so when Kanalog puts out 10V it provides the maximum torque you will ever need. But for now I would at least make sure they are set similarly

    Regards
    TK http://dynomotion.com


  6. #26
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch0-2-jpg
    minus gain input


    prevous
    servo madness-ch0-1-jpg

    Attached Thumbnails Attached Thumbnails servo madness-ch0-1-jpg  
    Last edited by jensenjim; 02-25-2019 at 02:39 PM. Reason: wrong picture


  7. #27
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4043
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    Hi jensenjim,

    so minus gain on input flipped the motion direction
    Well true but no. You already verified the encoder (Input) direction was correct. So changing the Input Gain would cause it to appear to move in the correct direction when it is in fact still physically moving in the wrong direction. So instead the Output Gain should be changed as described in my previous post.

    Regards
    TK http://dynomotion.com


  8. #28
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    lol magic voodoo electro tech guru language sure felt good for a second will continue



  9. #29
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch0-3-jpg

    isn't this 700 counts in the plus direction, I mean this looks like the correct



  10. #30
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch0-4-jpg

    when i change the output gain in kmotion it just increases the power of the servo but looks the same



  11. #31
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    i am still trying to understand your prvous post, under
    (do I still need to work with the servo drivers or can this be corrected with the kmotion)



  12. #32
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch0-5-jpg

    ok blue line 500 comes from move input
    right side scale is max output 200
    red is the result of changing output from servo



  13. #33
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch0-6-jpg

    when i increase it to 2000 output it is over powered, the unit jumps hard
    i understand its almost max power 2047 being 10 volts



  14. #34
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch0-7-jpg

    i expected the blue line to match the red
    i raised the step to 1700



  15. #35
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo 1 is moving in the wrong direction looking for a servo driver solution



  16. #36
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    reversed the drive
    Attachment 413396


    thats 8 pictures, reminds me of a bad rotor balance day at work, 8 starts , boss gets steemedservo madness-ch0-8-jpg



  17. #37
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    as far as the direction goes this is what is

    when i pulled the gantry , which has 2 servos attached ch0 and ch1 both axis moved in the same direction.
    when I ran the motor in console with dac 200 for both they both went clockwise which put ch 0 movint toward and ch 1 away. I trired to work with the driver but was unable to reverse the servo so I switched the output wires on the driver.



  18. #38
    Member
    Join Date
    Jun 2013
    Location
    USA
    Posts
    1041
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    You are missing a lot of what's being said. I think going by the amount of posts your trying hard. You should probably stop testing the motor for now. You instead need to use the help files in kmotion as a library. When someone suggests you change specific settings in certain ways your pictures show you are not doing so. I feel that's because you don't really understand the terminology so your skipping those suggestions. I was the same way when I started doing these types of things. A little reading of the first section of the help file in kmotion labeled kmotion/kflop executive software screens will save you a lot of headaches getting started.

    As for changing the gains on the configuration and flash screen. I noticed that in different photos you have different values on the gains. Presently you have output channel 0 set to .5. In a different picture you had channel 0 input gain set at 2. Now it is 1. The input channel gain and output channel gain can be positive or negative 1. Do not use any other values. Doing so I believe would cause a major issues between axis.

    Ben

    Sent from my E6810 using Tapatalk



  19. #39
    Registered
    Join Date
    May 2008
    Location
    usa
    Posts
    32
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    servo madness-ch0-9-jpg



  20. #40
    Member
    Join Date
    Jun 2004
    Location
    Scotland
    Posts
    355
    Downloads
    0
    Uploads
    0

    Default Re: servo madness

    Couple things that stand out, set the Feed Forward values to zero for now. Feed Forward should only be used to home in on a final tuning (they essentially try and predict what the axis is going to do, and although they will appear to give you a better tune and mask a poor tune, it can give you problems).

    As both your encoders do count the same way, they're fine. Leave your input gain at 1.
    As long as the drive can handle the +/-10V wires being reversed, it'll be fine now. The easier way would of been to set the output gain to -1.

    With that done, and with your motors still uncoupled, concentrate on a single channel/motor.
    Gradually start increasing P (0.5 at a time should be pretty safe), then do a Move (500 is a good starting point). What you're looking for is the difference between the command and position, and how stable/smooth the output and position is.
    Keep increasing the P and doing Moves until you notice oscillation on the output, at which point start increasing D (you can probably increase this by 1 at a time).

    Keep repeating the above, until you reach the point at which you can't get the value any higher without causing the output to oscillate.

    Then add in a bit I. Start with 0.001 increases, and if it's not doing much after a couple increase, jump to 0.01. Depending on the system, a little bit I can cause things to go unstable very quickly, whereas other systems will need quite a bit I to go unstable.
    Once things start to go unstable, lower P and D to regain stability.

    Only do this a couple times, as you're only aiming for a basic tune to ensure you've got reasonable stability at this point.

    Once that's done, increase your Maximum Output Limit to 2047, the run another 500 move to ensure things are still stable.
    Now you want to start finding the limits of the system, so start increasing the move size.
    It's probably worth doing some maths at this point to establish what you want your maximum motor speed to be, as there's not really any point tuning the motor above a speed they're ever going to run.

    Once you've established your maximum motor speed, and adjusted the tuning if needed by repeating the above, copy all the settings across to the second motor, and do some moves to ensure it's response is the same.

    Now I need to Tom to clarify how the KFlop handles slaved axes at this point. If they're slaved, does the slaved axis follow the master axis when using the Step Response screen?
    If the slaving remains active, I'd say couple the motors up, and then work towards a final tune.



Page 2 of 3 FirstFirst 123 LastLast

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

servo madness

servo madness