Welcome to Our Community

Some features disabled for guests. Register Today.

Suggestions to streamline my router motors?

Discussion in 'CNC Mills/Routers' started by JP Warnke, Sep 10, 2017.

  1. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9
    I recently built a CNC router with the help of my father: http://openbuilds.com/builds/warnke-cnc-router.4383/ . We originally intended to start out with something smaller and we just ended up wanting to make a larger router (Im a machine designer by trade and my father is a machinist who just sold his company, so we wanted to do something a little ambitious).

    (X) 4’ wide x 2’ deep with 8” z travel. X/Y/Z all run with Nema 34s (Y is dual driven). Ball screws and hiwin lin rails/bearings everywhere. It moves nicely and holds position well. The first cuts looked nice. Speed though is a little slower than we’d like – it will handle about 200 ipm rapid (trips if we run faster) and seems to feed just fine at speeds nearly just as fast if I run the spindle up high (18000 max rpm).

    I had a couple questions for anyone here who may know:

    1. On the Z motor - we ran an hour long cycle on the first test part (just cutting a simple 5/8” deep x 6” X 8” relief). The Z motor was naturally busy the whole time. At cycle’s end the z stepper was pretty hot - you could still grab it, but it was quite toasty.

    I’ve read that some of these steppers can heat up to 90-100deg Celsius before you’ve overplayed your hand, but I want to be able to cut large reliefs on this table; so I want to know I can run the router for 12-16 hours at a time now and then. Is the solution a better motor? Is the solution to change the electronics/settings around? As a mechanical guy with little electrical experience my first though is "what if I just counterweight my spindle assembly?" Right now the Z motor has to lift and lower about 25 lbs rapidly for any kind of relief cutting. If I put a 20lb counterweight on it could I improve that?

    2. I just want to get this all faster. 200 ipm isn't bad I guess. I was reading the specs for the camaster stinger (which also uses nema 34s) and they are getting much faster rapid speeds; I'm wondering why that is? What about that machine enables them to utilize the motors more effectively?

    I am pretty certain that the answers to both of these may be electrical (My weakness is the electrical side of all this). Any ideas/suggestions?

    Thanks all

    -JP
     
  2. Gary Caruso

    Gary Caruso Veteran
    Builder

    Joined:
    May 19, 2016
    Messages:
    219
    Likes Received:
    83
    Nice build don't know how i missed this one! I scanned through the build but didn't see what voltage you are running? Voltage is speed,, so if your drives can take more voltage upgrade your PSU to higher voltage. Also if you are running high micro steps, lower them so its fewer pulses per turn. 1/4 Micro stepping should be fine for screws.
    As for the Z getting hot, easy way is to reduce the current setting one notch lower or add a cooling fan. After an hour you probably maxed out, so a fan would be fine IMO. Another option is a linear spring.
    Adding counter weight is a problem of adding mass which you then have to accelerate and decelerate, which will force you to lower your acceleration setting.
     
  3. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9
    Thanks. It's been a long learning process! (about a year now)

    Ill try not to botch my explanation on electrical stats (I'm rather a dumb dumb when it comes to electricity - my dad is really the one who read up and designed and built the controls side of this) - Our motors are 1200 oz-in. I know we are running on 220V and each motor power supply is set at 220V (there is a switch for 220 or 110), though all that gets converted to DC by the drivers.

    I know that we do have our microsteps set rather high right now; that was our solution to getting some of the vibration out - between 10 and 70 ipm travel we were getting some kind of acoustic vibration in the Y-dual driven travel when we first ran it and upping the microstepping seemed to help that significantly; we only really hear anything at this point between 10 and 20 ipm travel.

    On the Z...good...I don't want to add a counterweight anyway; itll look dumb, hah X-D

    So are you saying an extension spring? I thought about springs; compression springs are nice to deal with but I have a lot of travel (8") so id have to do something funky to get a compression spring with that kind of range; but even so, I wasn't sure how that would play out - since resistance on the ball screw would then fluctuate depending on position of the spindle up and down - in the case of an extension spring, further out would mean reduced weight hanging from the ball screw; but further up would fail to do much of anything. In the end it seems like it shouldn't make much difference; it's a 5mm pitch ballscrew, so ive got to believe its not that taxing to hang 25lbs or so of aluminum plate and spindle and such off of it. If I can avoid adding any kind of mechanical contraption, I will do that.

    I think I probably just need to go for it and run a long cycle to see what happens. The motors were pretty cheap (at $50 a piece I won't cry if I lose one). If the motor overheats; it's on a 3amp breaker; so ...the only thing I think Id have to worry about is my other axes continuing to move...which brings up another question I had -

    Any idea if there is a way to stop a machine cycle when I have a breaker trip? In special machine design, they use a PLC to control all that stuff; but I don't have the money to drop for a PLC (or the time to read up on programming one). Is there a trick I can use to freeze cycle if I have a breaker trip/motor fail?
     
  4. Gary Caruso

    Gary Caruso Veteran
    Builder

    Joined:
    May 19, 2016
    Messages:
    219
    Likes Received:
    83
    I read through your discussion a little and i see this?

    - (5) NEMA 34 640 oz-in Stepper motors (KL34H280-45-8A)
    - (5) KL-5056D Digital Bipolar Stepper Motor Drivers - 32 bit DSP based
    - (6) Motor cable connectors
    - (1) Handwheel MPG (wireless) for Mach 3 controller, 4-axis (X, Y, Z, A)
    - (1) Water Pump 200 GPH for spindle cooling system
    - (1) 2200w 3HP Water cooled CNC Milling Spindle,KL-2200
    - (1) 2.2Kw (3HP) VFD for Spindle (KL-VFD11) 110VAC input, Connect R and T
    - (1) Unregulated linear 960W/48 VDC/20A Toroidal PSU (KL-4820) with 5VDC
    - (1) CNC 6-Axis interface breakout board with relay and spindle control
    - (1) 24V 2A Switch power supply.


    if you are still using those drives then you have an ac to dc power supply in there somewhere, those drives don't run on 220AC.
    In one pic of your build i see a PSU but its a switcher, not a big 20A Toroidal PSU.. looks like the drives can do 45vdc

    If you are running high micro-steps then that would be why you can't get more speed..but 200 IPM isn't too bad..

    I really don't know why you would run a breaker on each motor, IMO this is not needed. Steppers are not like brushed motors, so when stalled you will not burn a winding like on a brushed motor.
    they are drawing just as much current holding position as when moving (until your drive throttles back when not commanded for some time).

    If things have changed for hardware a bit since the build thread update them and get some pics of your control box, this would help with the advice. :)

    Sorry not linear spring, my bad, "constant force spring" commonly used for things like this.

    Cheers
     
    JP Warnke likes this.
  5. piter

    piter Well-Known
    Builder

    Joined:
    Sep 11, 2014
    Messages:
    6
    Likes Received:
    2
    For managing heating issues, plug one (or even two) cpu heatsink with fan on a side of the stepper motor: I did it on my CNC mill and worked perfectly
     
    JP Warnke and David the swarfer like this.
  6. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9
    IMG_2275.JPG
    IMG_2277.JPG IMG_2278.JPG IMG_2279.JPG IMG_2280.JPG IMG_2283.JPG IMG_2284.JPG IMG_2285.JPG IMG_2289.JPG

    Sorry. Busy work week (nearly 80 hours ....job is getting to be too demanding).

    Anyway; some of the components I have changed from my original scope.

    Power supplies are 60vdc I believe.
    Using air cooled 3.3 kw spindle now (not water).
    All motors are nema 30, but like 1200ish oz-in I believe.

    The motors definitely heat up when I'm doing big relief carvings. I ran it for about 2 hours the other night and z motor was warm enough I could only keep my hand on it for about 15-20 seconds.

    More importantly than that though I noticed I am gradually losing my z position when cutting a relief. I started running a relief I generated in artcam and noticed that about an hour or two into it on the finishing cut I had walked upward (not down) about .03". Definitely didn't push the tool up (I checked). It is cutting left-right finish passes from bottom to top. Takes about an hour to move up a couple inches. It actually lost zero somehow. I don't know if the frame is too complex or what? I ran it again and same result. It seems to be a gradual creeping upward due to a lot of rapid motions up and down in z. Like it's missing a few steps here and there but moving upward gradually every time. Makes no sense .

    IMG_2317.JPG
     
  7. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9
    Note: most of the garbled stuff on the top half of the relief is my roughing pass. I just stopped cycle before cutting up anymore than 2" or so. It is an stl of a deer relief with a frame.
     
  8. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9
    I'll have to try that. Thanks!
     
  9. Gary Caruso

    Gary Caruso Veteran
    Builder

    Joined:
    May 19, 2016
    Messages:
    219
    Likes Received:
    83
    If the Z is getting that hot, set the current one step lower on the drive switches, maybe your acceleration is too aggressive on the Z causing the drift. or a setscrew on the coupling is loose.

    BTW that is one awesome machine!
     
    #9 Gary Caruso, Sep 19, 2017
    Last edited: Sep 19, 2017
    JP Warnke likes this.
  10. David the swarfer

    David the swarfer OpenBuilds Team
    Staff Member Moderator Builder

    Joined:
    Aug 6, 2013
    Messages:
    784
    Likes Received:
    362
    creeping Z 'UP', very weird.

    if it was missing steps because of mechanical binding, or too high acceleration, then Z would drop rather than rise.

    so it is something else....... and my idea is this...
    some drivers need a delay between step pulses and direction changes, and if those delays are too short it might well do an extra step during the direction change. so if your CNC controller has settings for step/dir change delays, increase them and see what happens.

    EDIT
    check the timing diagrams at
    http://www.leadshine.com/UploadFile/Down/MA860Hm.pdf
     
    #10 David the swarfer, Sep 19, 2017
    Last edited: Sep 19, 2017
    JP Warnke likes this.
  11. Gary Caruso

    Gary Caruso Veteran
    Builder

    Joined:
    May 19, 2016
    Messages:
    219
    Likes Received:
    83

    very true about the creep up vs down especially with the gravity of his spindle! Could the timing thing be helped if they went from 1/16th to 1/8 micro stepping? I believe all his drives are set for 1/16
     
  12. David the swarfer

    David the swarfer OpenBuilds Team
    Staff Member Moderator Builder

    Joined:
    Aug 6, 2013
    Messages:
    784
    Likes Received:
    362
    that would make no difference at all as it does not affect the timing of the pulses relative to the direction pin change.
    the timing diagrams indicate a minimum delay of 5us after the direction pin changes before the next step pulse (for that driver).
     
  13. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9
    Thanks! My family owned a machine shop for a bit over 30 years and sold it. This was a retirement project my dad and I decided to try our hand at. I did the mechanical design in SW and he did the controls design and most of the machining on his Bridgeport. I have to check my timetable but seems like it has been about a year since I started the design now.

    I still need to design up a rotary axis and a shift fixture with some dead leg stands for front back of the machine. I originally wanted to keep this small for the garage; but still be able to carve doors and some decent sized round pieces (maybe 7" or 8" dia max?).

    It's been a fun project and we have both loved learning as we go!
     
  14. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9

    David, thanks! This is one thing pops says he hasn't tried yet; he will look into it tomorrow!

    He did also just say something about switching a voltage setting on the PMDX (sounds like it has a setting that can either be 0-5V or 5-0V???). I'm not sure what he meant by that but I guess when he switched it to the other setting and reran the cycle he noticed that it cut position loss in half from what we had originally been seeing.

    Sounds like he also lowered current some like Gary suggested to try to mitigate some of the heating we were seeing. So we will see if that helps.

    Thanks everyone for the input. I really appreciate it!
     
  15. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9
    So my father did look into this. What I understand is that the drivers we have aren't actually leadshines; they are imitations. Evidently the jumpers shown in this leadshine manual do not exist in the drivers we have.

    I'm told changing some items and modifyng parameters did affect the operation of the Z motor, but each change either did nothing or made it worse. In the end he went back to the original settings which yielded the best results; but that still meant a deviation of about .01" shift upward per inch of z-carving at 100 in/min feed rate.

    I wonder if it would be worth buying the actual leadshine to swap out and try?

    I confess, I still don't understand this much myself; I'll have to read up when I get some time. Appreciate the help though!
     
  16. David the swarfer

    David the swarfer OpenBuilds Team
    Staff Member Moderator Builder

    Joined:
    Aug 6, 2013
    Messages:
    784
    Likes Received:
    362
    You need to change the timing in the controller. Nothing changes in the drivers themselves.

    In other words.... your z is stepping along in the up direction, stops, change direction, and now stepping down starts.
    The driver requires at least 5us delay between the changing of the direction pin and the first step pulse. (If they are at least similar to the documents I linked).

    To set that delay you have to change the controller setup since the controller is sending the pulses.
     
    #16 David the swarfer, Sep 24, 2017
    Last edited: Sep 24, 2017
    Gary Caruso likes this.
  17. Gary Caruso

    Gary Caruso Veteran
    Builder

    Joined:
    May 19, 2016
    Messages:
    219
    Likes Received:
    83
    Could that explain his motor heat issue too?
     
  18. David the swarfer

    David the swarfer OpenBuilds Team
    Staff Member Moderator Builder

    Joined:
    Aug 6, 2013
    Messages:
    784
    Likes Received:
    362
    No, step timing does not affect heat.
    All motors generate heat. You are putting power through them.

    Example numbers.....
    24volts times 2amps is 48 watts. A car headlight bulb is usually about 55 watts and will burn you instantly. Little wonder that the motors warm up a bit . If you can touch a motor for a few seconds without getting burnt they are fine. Most are rated up to 80C for normal operation, and 80 will burn you.

    If the heat really bothers you can reduce the amps and increase the volts but I would much rather add some fans.
     
  19. JP Warnke

    JP Warnke Journeyman
    Builder

    Joined:
    Jan 4, 2017
    Messages:
    32
    Likes Received:
    9
    Well, we've changed this; our motion controller has delay settings for either 5, 10 or 20 microseconds and we've tried them all. Running at 10 seems best but still creeping up about .005" per inch of travel in one axis for every 12inches of Z carving in the perpendicular axis. It's a very consistent position loss and creep upward (really weird)

    We are going to test run a simpler relief to see if we get the same results.

    To make things even more interesting our variable spindle is now having issues turning off. You'll tell it to shut off in Mach 4 and it only slows slightly...sometimes it stops after a few minutes....sometimes not.

    I think at day's end I suspect we will find we have made a mistake getting Mach 4 and the pmdx controller we chose.
     

Share This Page

  • About Us

    The OpenBuilds Team is dedicated helping you to Dream it - Build it - Share it! Collaborate on our forums and be sure to visit the Part Store for all your Maker needs.


    [email protected]

  • Like us on Facebook

  • Support Open Source FairShare Program!

    OpenBuilds FairShare Give Back Program provides resources to Open Source projects, developers and schools around the world. Invest in your future by helping others develop their future.

    Donate to Open Source