Options to drive 24vdc motor - Page 2


Page 2 of 2 FirstFirst 12
Results 21 to 31 of 31

Thread: Options to drive 24vdc motor

  1. #21
    Member
    Join Date
    Apr 2007
    Location
    Canada
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Only took me 2 years to get around to this...Spent the evening tuning the motor. Right now I have it disconnected from the slew bearing, but the planetary gearbox is still attached to the motor. There's no easy way to disconnect it, as the motor and gearbox are in a riveted housing.
    To summarize:
    I'm using the AMC 12A8 brushed servo drive I got from Al.

    Options to drive 24vdc motor-amc12a8-jpg

    It is operating in Current (Torque) mode.

    I'm using a single ended to differential converter from CNC4PC, I seem to be getting 8 counts per motor revolution. So for 1 rev on the output of the planetary gear I get (234:1) 1872 counts. This is reduced further by the slew drive (78:1), so will have 146,016 counts per revolution of the axis.

    Here's a video just showing the settings, plots, and a move. The first plot/move is done with velocity set at 240 (1872 counts per rev of planetary output X 7.7 revs per minute / 60), this was the end result after basically just increasing P gain. Next I increased velocity to saturate the output (then backed it off a bit) and played around with feed forward. Both have a 2nd order low pass filter of 500Hz Q=1.4. Increasing D gain didn't seem to have much effect.



    I think these look pretty good? Or no?

    Only thing I wasn't sure about was what do with the settings on the servo drive. It is setup to operate in current mode. To set it up in velocity mode would require a tach, the manual suggests that current mode is the preferred mode anyway. There is one on board potentiometer to set something called Reference gain:

    Reference Gain
    This potentiometer adjusts the ratio between the input signal and the output variable (voltage, current, velocity, or duty cycle). For a specific gain setting, turn this potentiometer fully counter-clockwise, and adjust the command input to 1V. Then turn clockwise while monitoring motor velocity or drive output voltage (depending on mode of operation) until the required output is obtained for the given 1V command. Turning this potentiometer counter-clockwise decreases the reference in gain, while setting this potentiometer in the fully clockwise position makes the whole range of drive output available. This potentiometer may be left in the fully clockwise position if a controller is used to close the velocity or position loops.

    I have it in the fully clockwise position.

    Tomorrow I'm going to attach the motor/planetary gear to the slew gear and hopefully the tuning doesn't change much. The good news is that I have the rest of the fabrication work done. So provided that things still work ok once attached to the slew gear I can bolt everything together and try it a load on it. Just waiting on proper wire for the motor, should be here Monday.



  2. #22
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    3768
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Hi Greg,

    Those look reasonable but very difficult to see in a video. Please post settings and plots instead.

    Looks slow. Do you have speed requirements?

    You might test how many counts to get a full revolution of the shaft.

    You might calculate the motor RPM to see how it compares to expected motor RPM for that motor.

    Normally the reference gain of the amplifier should be set such that when 10V is commanded the amplifier drives at slightly more speed or torque than is ever needed. I suspect that although configured for current/torque mode it is really operating in more of a voltage/speed mode using the back-emf of the motor to control speed, otherwise it would be unstable without D gain.

    Regards
    TK http://dynomotion.com


  3. #23
    Community Moderator Al_The_Man's Avatar
    Join Date
    Dec 2003
    Location
    Canada
    Posts
    23687
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    The AMC instructions for the 12A8 specifies setting the pot #1 full CCW when used in Current mode use.
    Pot 2 is current limit
    Pot 3 reference gain, adjusts ratio between input signal and output variable (voltage, current or velocity)
    Pot 4 offset/test only relevent when dip switch 4 is on.for testing

    CNC, Mechatronics Integration and Custom Machine Design

    “Logic will get you from A to B. Imagination will take you everywhere.”
    Albert E.


  4. #24
    Member
    Join Date
    Apr 2007
    Location
    Canada
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Quote Originally Posted by TomKerekes View Post
    Hi Greg,

    Those look reasonable but very difficult to see in a video. Please post settings and plots instead.
    Hi Tom,
    Yes sorry about the video, pause helps.
    It is slow! But that is OK, it is supposed to be slow. This is for the stone saw, with feed rates between 0.25" to 1.5" per minute on the XY axes. So far most of my cutting has been around 1" per minute. I'm still trying to wrap my head around all things rotary, but I think the axis should be plenty fast enough for cutting. Positioning (G1) moves, well they are big pieces of stone, probably better for safety that the axis can't move all that fast.

    The counts I gave above seem to check out. At 240 counts/sec, it takes approximately 10 minutes to complete one rev of the big gear, which is right at the spec of 0.1 RPM. I used that logging/plotting software I wrote a while back today and did some longer moves and was able to see OK performance up to 320 counts/sec. If I try to go higher the 24VDC power supply faults out. Kind of weird as it doesn't trip the breaker I have between it and the drive (4 amps) or blow the 2 A fuse on the input to the power supply. It just faults and shuts down. There is a potentiometer to play with to limit the servo drive output, I have it set where it should be limiting constant to 4 amps. It trips right when it starts the second move. Anyway 320 counts/sec it is taking about 7minutes and 40 seconds for 1 rev, so that's pretty good.

    Is there a standard way to measure backlash in a rotary axis? I know how to do it on a linear axis. I used a similar method:

    Not sure I'm measuring at the correct location.

    Thanks.

    Here's some plots at 320 counts/sec, with motor attached to the big slewing bearing. Second one is zoomed in.

    Options to drive 24vdc motor-axis-320-countssec-velocity-jpg

    Options to drive 24vdc motor-axis-320-countssec-velocity_zoomed-jpg

    Options to drive 24vdc motor-velocity320_settings-png



  5. #25
    Member
    Join Date
    Apr 2007
    Location
    Canada
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Hi Al,

    Yes I have all the settings in the manual for current mode. Thanks!



  6. #26
    Member
    Join Date
    Apr 2007
    Location
    Canada
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Positioning (G1) moves, well they are big pieces of stone, probably better for safety that the axis can't move all that fast.
    Meant to say G0 moves.



  7. #27
    Member
    Join Date
    Apr 2007
    Location
    Canada
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Hi Tom,

    I read up on the wiki for backlash. I used the method of moving small amounts until the my dial indicator read zero to find how many counts of backlash. I did this at 4 points (the gauge was left in the same place, the gear was rotated). I recorded the amount of counts it took in each direction.

    Code:
    Position	Pos Backlash Inch	Pos Backlash Counts	Neg Backlash Inch	Neg Backlash Counts
    1	        0.006	                22	                                0.006	                22
    2	        0.009	                28	                                0.009	                28
    3	        0.01	                    28	                                0.012	                40
    4	        0.011	                35	                                0.014	                40
    Average		                        28.25		                                                32.5
    I have 146,016 counts/rev of the gear, 405.6 counts/degree. Do you think it's worth using the backlash compensation to correct? Or am I over analyzing this? I did apply it in the motor setup screen and it seemed to work.
    Thanks

    Options to drive 24vdc motor-backlash-jpg



  8. #28
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    3768
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Hi Greg,

    The servo errors seem very small compared to the backlash (3 counts vs 28 counts) as expected for such a mechanism.

    The backlash seems inconsistent so will likely to be difficult to correct. At best possibly reduce it by half?

    I think of backlash like trying to pull/position a sled with a chain. If you go slowly (your case) and there is lots of friction (not sure) or preload (gravity?) and small external forces it may work well. It’s not clear to me how your stock will be mounted and such. If the stock is mounted lopsided and the direction of gravity shifts that may be a problem.

    Assuming a backlash of 28 I calculate that translates to approximately +/- 1/64th of an inch at a radius of 2 feet. Is that tolerable for you?

    Regards
    TK http://dynomotion.com


  9. #29
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    3768
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Oops

    Last edited by TomKerekes; 05-04-2021 at 10:15 AM. Reason: Double posted
    Regards
    TK http://dynomotion.com


  10. #30
    Member
    Join Date
    Apr 2007
    Location
    Canada
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Hi Tom,

    Thanks for the insight. +/- 1/64" is great, probably not worth worrying about, but I'll still try out the backlash compensation and see what I observe.

    FWIW this is how the gear mounts.
    First picture has the table that rotates hidden to show the gear.

    Options to drive 24vdc motor-tablegear-jpg

    Options to drive 24vdc motor-table-jpg

    If I understand what your saying, I should take some backlash measurements loaded to confirm that the gear behaves similarly as when it is unloaded. I'll do that.

    Thanks!



  11. #31
    Member
    Join Date
    Apr 2007
    Location
    Canada
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default Re: Options to drive 24vdc motor

    Hi Tom,

    Had a small setback. The servo drive failed. Not sure why but it fails to power up. Luckily there are a lot of these AMC drives on ebay. I also learned that you can use all the their analog brushless drives on brushed motors. Ended up with a B12A6 model. I started over and followed the wiki more closely this time. I know it's all kind of pointless on this machine, but I feel I understand the process a bit better now.

    2 Order moves:
    320 counts/sec works out to ~ 47 degrees/min.
    I couldn't go higher than 1200 (max v in 0.27sec) for acceleration or the current draw was more than the power supply could handle. It's highly doubtful that I'll ever be cutting at these high of speeds. But I do have a better power supply on the way.

    3 Order moves:
    Reduced acceleration a bit to 1000 and ended up with jerk set low to 1000. The improvements can be seen on the plots over the infinite jerk 2nd order plots, but real world will never see it. Don't really need the higher acceleration for positioning the stone so I may reduce this.

    I'm curious how it will perform once loaded. I may have to redo all this.

    I added a slope feature to the plotting tool. In these plots I have it configured so that it does the delta between 10 readings, otherwise the plots for v/a/j are too hard to see the trend as the instantaneous values create a "spikey" plot I also hooked up the current reference from the servo drive to an analog input and is shown in the plots.

    Attached Thumbnails Attached Thumbnails Options to drive 24vdc motor-axis-2ndorder-320v1200a350000jp120i0-01d4000ff0-6_pos_err_amps-jpg   Options to drive 24vdc motor-axis-2ndorder-320v1200a350000jp120i0-01d4000ff0-6_pos_err_amps_zoomed-jpg   Options to drive 24vdc motor-axis-2ndorder-320v1200a350000jp120i0-01d4000ff0-6_vajc-jpg   Options to drive 24vdc motor-axis-2ndorder-320v1200a350000jp120i0-01d4000ff0-6_vajc_zoomed-jpg  

    Options to drive 24vdc motor-axis-3rdorder-320v1000a1000jp120i0-01d4000ff0-6_pos_err_amps-jpg   Options to drive 24vdc motor-axis-3rdorder-320v1000a1000jp120i0-01d4000ff0-6_pos_err_amps_zoomed-jpg   Options to drive 24vdc motor-axis-3rdorder-320v1000a1000jp120i0-01d4000ff0-6_vajc-jpg   Options to drive 24vdc motor-axis-3rdorder-320v1000a1000jp120i0-01d4000ff0-6_vajc_zoomed-jpg  



Page 2 of 2 FirstFirst 12

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

Options to drive 24vdc motor

Options to drive 24vdc motor

Options to drive 24vdc motor