Page 2 of 7 FirstFirst 1234567 LastLast
Results 16 to 30 of 98
  1. #16
    Join Date
    Sep 2012
    Location
    York, North Yorkshire UK
    Posts
    6,439

    Default

    Quote Originally Posted by neevo View Post
    Thanks gents! What I needed regards to info and also thanks for the link to the Clough42 questions section (Nigel), I hadn’t seen that before.

    I copied Clough42 a bit too much and I’m hindsight I would have gotten the 1000 encoder vs the 1024. But I’m sure I’ll get it working with minimal error, certainly over a 1-2 inch thread.
    Hi Neevo,

    I think that the 1024 encoder is right, because an eight TPI leadscrew will divide equally into 1024.
    1024/8 = 128. Each leadscrew turn would equate to 1/8" of an inch, 125 thou or 3.175 mm.
    Best Regards:
    Baron J.

  2. #17
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default

    Thanks Baron, sounds like I might have mistakenly got it right then! Can’t wait to get all my parts to give it a bench test!

  3. #18
    Join Date
    Aug 2019
    Location
    Revesby - Sydney Australia
    Age
    56
    Posts
    1,183

    Default

    Quote Originally Posted by BaronJ View Post
    Hi Neevo,

    I think that the 1024 encoder is right, because an eight TPI leadscrew will divide equally into 1024.
    1024/8 = 128. Each leadscrew turn would equate to 1/8" of an inch, 125 thou or 3.175 mm.


    Neevo, any encoder with a highish ppr count will be fine,
    because the software will scale the ratios better than our brains can




    Baron, not sure it is that simple
    because the spindle and its encoder are "driving" the pulses and math.




    Note that I haven't looked at the source code of his software, and this may all be moot,
    but consider this thought experiment...


    - Lets imagine we are cutting 8tpi for a new backplate/faceplate/collet chuck, and need 1:1;


    1. Spindle turns 90°. Quarter turn causes 1024ppr encoder to send 256 full pulses,
      or 1024 quadrature-edge pulses
    2. Software multiplies by 0.39 or 0.391, and divides by 8tpi,
      generating 49.92 or 50.048 motor pulses
    3. 200 step/rev motor turns the feed/lead screw turns either 49/200 or 50/200 of a turn
      (assuming no motor step down pulleys)




    - Let's consider a smaller rotation of the spindle;


    1. Spindle turns 1/16 revolution, or 22.5°, which causes encoder to send 256 q-e pulses
    2. S/w multiplies by 0.39 or 0.391, divides by 8, generates 12.48 or 12.512 pulses
    3. Screw turns by 12/200 of a revolution, or 21.6°





    So, small or slow spindle revolution, or large pitch threads, can cause growing errors.


    There are many software tricks that can correct this; "successive approximation registers", more accurate floating point constants et c. (and I suspect Clough's system does better than this simplistic input pulse mapping maths)


    There are also hardware ways to minimise this, like higher ppr count on the encoder, higher step count on output motor.





    Practically speaking, though, we often cut threads with high TPI, or Metric on an Imperial screw, or matching strange/odd/prime standards, so the spindle to screw ratio is never tidy,



    but for bench testing a stepper motor system, having perfect integer mapping of input to output devices makes life simpler!







    Now, if only I could justify buying a few "smart motor" devices for all my lathes and mills

  4. #19
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default Clough42 Electronic Leadscrew Implementation

    I think the stepper is actually set for 200 pulses per revolution but also with micro step functionality. So would that improve the resolution to the point where part steps are a non issue?

    I then also have the option to adjust this in the software too.

    So 4096 pulses per rev on the encoder. Can’t mess with this because the spindle RPM won’t read properly.

    Set up the leadscrew as 8 TPI (I assume this should drive the conversion factor from the encoder).

    Configure the motor setup for 200 pulses per rev with 16 micro steps (1600 per rev).

    Set up the motor controller for 200 pulses per rev with 8 micro steps (to account for 2:1 ratio on final drive).

    I should be able to set the controller to 8tpi and see it turn at half the rate of the encoder.

  5. #20
    Join Date
    Aug 2019
    Location
    Revesby - Sydney Australia
    Age
    56
    Posts
    1,183

    Default

    Hmmm. Microstepping. Didn't know about that. A quick Google showed a few pages that just confused me

    e.g.https://www.linearmotiontips.com/microstepping-basics/

    but I did pick up that using it lowers torque to about 70% ?




    If you are gearing stepper down 2:1, motor is effectively 400 steps/rev.

    Multiply by 8TPI. That gives you 3,200 steps per inch.
    Each step moves the carriage 0.0003125 inches.
    About 8 microns or a third of a thou - plenty of resolution for me

    I would just configure as 400 steps/rev with no micro-steps.
    Last edited by nigelpearson; 26th Aug 2021 at 04:26 AM. Reason: Calculate full step resolution

  6. #21
    Join Date
    Aug 2011
    Location
    Melbourne
    Posts
    4,779

    Default

    My one and only nema24 stepper motor is 1.8 degrees per pulse. That's 200 steps per revolution. However, stepper drivers can be configured to deliver 1/2 steps, (400 per rev) 1/4 and 1/8 steps by partially energising multiple phase windings. As mentioned, partial stepping reduces torque but this may not be an issue.

    These configurations would need to be set up somewhere in your settings, so that it knows by how much your stepper moves per pulse and therefore can calculate how many pulses to deliver per rpm based on the pitch selected.

    Simon

    Sent from my SM-G970F using Tapatalk
    Girl, I don't wanna know about your mild-mannered alter ego or anything like that." I mean, you tell me you're, uh, super-mega-ultra-lightning babe? That's all right with me. I'm good. I'm good.

  7. #22
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default

    Small parts came in today. My GT2 pulleys which I need to modify to attach to the change gear main spindle and also one on the encoder.




    The bores are 8mm but I need 14.32mm for the lathe which means I’m going to mostly turn off the shoulder on one, bore out to the correct diameter, then try and fashion a key way slot.

    It will be held on the stud with the existing nut:



    Once the shoulder is turned off, it will be flush with the end of the start of the thread.

  8. #23
    Join Date
    Aug 2019
    Location
    Revesby - Sydney Australia
    Age
    56
    Posts
    1,183

    Default

    I can't visualise the next stage. How you are going to attach the gear to the "change gear spindle?"
    (which I think is actually called the "Stud Gear?")




    P.S. that gear is smaller than the gear on the spindle. On my Hercus, it usually has a 16tooth on it, but the spindle has something like 26 teeth machined into it?

    That hugely affects the pulse count from your encoder.

  9. #24
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default

    I’m going to bore out the pulley to match the stud gear and mount it in place of the gear.

    I checked and my spindle is something like 26t, then the reverse tumblers are something like 36t and then the back of the stud gear is 26t again. So the ratio on the stud gear is the same as the spindle (that I confirmed before I ordered the pulleys).

    I have the pulley in the chuck already and plan to bore it out tonight.

  10. #25
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default

    I checked the spindle and the gear is 26t:



    Then I confirmed that the tumbler gears are 32 and the gear behind the stud gear is also 26. So my plan to mount the pulley in front of this gear (where the stud gear goes), will yield me a 1:1 ratio with the spindle.

    I managed to turn the inside of the GT2 pulley, take off the flange and use a small end mill to notch out the key way (I need a better way to do this in the future). It’s on there nice and snug:



    I need to make a proper sized washer. But it will do for now. Here it is mounted back on the lathe:



    I’m planning on mounting the encoder close to the tumbler assembly and drive it off a belt:




    I’m really pleased with the alignment:



    So here are the bits so far, a nice and tidy package:



    And with the cover on, you’d never know what was in there:


  11. #26
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default

    Machined up an 8mm bolt for the encoder mount. Actually I did 2 because the first one I forgot to account for the mount and made the thread too short:



    Mounted up. I’ve ordered a 200mm belt to complete this part:



    Plus I managed to get all my electronics on a piece of melamine under the end of the lathe bench. Fits easily. The only change I might make is putting the TI board in a box, not really happy having all those delicate parts flapping in the breeze:


  12. #27
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default

    Whilst I have to bear the agonising wait for parts to come in. I made a couple of bits.

    First one was something that I’ve needed for a while: a depth indicator holder for the lathe. I had some old 3M Pulleys that weren’t being used, so they gave up their Alu.



    Here you can see the neo magnets I glued in to the base to keep it fixed to the lathe bed.



    The design was “fly by the seat of your pants” but I’m really pleased with how it came out. I made a brass thumb screw to hold the indicator but it’s really not needed as the fit is a really tight tolerance:



    On the lathe:



    The indicator registers on this flat on the carriage:



    Next job was to try and make something that will allow me to make a key way. My hack job on the encoder pulley was functional, but not exactly elegant. So I put some 5mm HSS in the mill and thinned it down with a carbide end mill.





    I have it set in a lathe tool holder and I’m using the carriage to cut the slot, 1 thousandth at a time, then pushing 3 passes. It works surprisingly well as I found out on a test piece of Alu:



    I need it to cut a key way in one of the pulleys for the leadscrew.


  13. #28
    Join Date
    Nov 2007
    Location
    melbourne australia
    Posts
    3,228

    Default

    Nice work on the DI holder. Do your DI a favour and fit a cap screw to the face of the holder that the DI probe protrudes from. It works as a bumper to prevent the DI being damaged if (when) you get distracted and drive the carriage hard into the holder.
    Chris

  14. #29
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default

    Quote Originally Posted by jack620 View Post
    Nice work on the DI holder. Do your DI a favour and fit a cap screw to the face of the holder that the DI probe protrudes from. It works as a bumper to prevent the DI being damaged if (when) you get distracted and drive the carriage hard into the holder.
    Excellent idea, will do, thanks for the tip.

  15. #30
    Join Date
    Sep 2009
    Location
    Sydney
    Posts
    666

    Default

    Really exciting. My leadscrew control panel and header board turned up from Clough42:







    Excited to get in the garage and solder this al together:



    Then I need to find a way to connect the control board to the TI prototype board and I can see if the code will load!

Page 2 of 7 FirstFirst 1234567 LastLast

Similar Threads

  1. Sieg C6 lathe - electronic leadscrew
    By snowyskiesau in forum METALWORK GENERAL
    Replies: 19
    Last Post: 29th May 2023, 06:04 AM
  2. Electronic leadscrew
    By phaser in forum METALWORK GENERAL
    Replies: 5
    Last Post: 4th Jun 2019, 10:48 PM
  3. Calling all electronic gurus
    By Karl Robbers in forum WELDING
    Replies: 20
    Last Post: 28th Jan 2016, 08:31 AM
  4. Servo electronic threading
    By localele in forum METALWORK GENERAL
    Replies: 19
    Last Post: 31st Aug 2012, 11:21 PM
  5. Where to buy electronic micrometers
    By steran50 in forum METALWORK GENERAL
    Replies: 4
    Last Post: 11th Aug 2012, 03:22 PM

Posting Permissions

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