584,862 active members*
5,722 visitors online*
Register for free
Login
Results 1 to 12 of 12
  1. #1
    Join Date
    Jun 2010
    Posts
    116

    ESS, Mach4, SGZH Servo Spindle Motor

    I'm in the process of rebuilding my Milltronics Partner 1 mill using Mach4 and ESS. I'm to the point of trying to get my SZGH servo spindle motor, https://www.automationtechnologiesin...-inverter-220v , wired in, but the inputs and outputs for the driver are different than what Mach4/ESS want. The attached screenshot shows what the driver is looking for and outputting. I've emailed SZGH and am waiting to hear back from them and Arturo at CNC4PC is looking into it as well, but I figured I'd see if someone here had any experience with this?

    Thanks,

    Mike

  2. #2
    Join Date
    Nov 2013
    Posts
    4282

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Hi,
    I have a 1.8kW Allen Bradley servo as a spindle motor and use it with Mach4 and an ESS matched to my own breakout board.

    Given that it is a servo there are a number of ways you can run it.

    The most likely way, and that apparently matches the pic you posted is to use pulse control. Most servos can be programmed to either Step/Dir OR CW/CCW OR quadrature....your choice.
    The ESS produces Step/Dir natively....so I would guess that what you would use. Note the the ESS outputs are +5VDC/0VDC logic, but the pic you posted suggests the input to the servo
    is differential. You'll need to convert the single ended logic of the ESS to differential input. You have not said what breakout board you are using? It may already have differential outputs.

    Another way you can use a servo is velocity mode with an analog +10VDC/-10VDC input. This is the mode I use the most. I use the spindle PWM output of the ESS to drive the servo
    with an analog voltage....just like a VFD....too easy!! The pic of your servo does not show that arrangement but I'd be surprised if it di not have it.

    Note that velocity mode would mean you don't have position control......if you want or need the spindle to index or you want a coordinated C axis for rigid tapping say, but for all other purposes
    analog voltage control is fine. You may notice on my breakout board I have a connection for a Caxis (Axis of Evil....I've had more cockups with the Caxis than all the others combined!!!).
    This means that I can hook up my servo driven spindle as a position capable axis for rigid tapping. It works, but I use it very seldom.


    Craig

  3. #3
    Join Date
    Jun 2010
    Posts
    116

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Thanks for getting back to me! Unfortunately, I will need orientation for tapping and for my ATC. It will accept +/-10V input and Modbus. I'm using a C82 BOB, this setup actually: https://www.automationtechnologiesin...-110vac-220vac . I'll have to see what Arturo comes up with, I think since it looks like I'll need to convert 0-10V to +/-10 or convert the EA+/- Eb+/- to step/dir. I'm still not sure how to get the quadrature encoder output from the drive into Mach. Another converter? These are things I should've looked into earlier, I guess . Hopefully Arturo can figure it out.

    Mike

    Nice BOB, BTW!

  4. #4
    Join Date
    Nov 2013
    Posts
    4282

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Hi,
    apparently plenty of spindles use an electromechanical spindle lock to stop the spindle at the correct orientation for tool changing.

    If you want position control then forget 0-10VDC or -10VDC/10VDC....neither offer position control, at least without a suitable motion controller, and the ESS is certainly not such a device.

    Converting Step/Dir to EA+/- and EB+/- is easy, just a line driver IC, a 26LS31 is the industry standard go-to quad line driver. I used AEIC7272 quad line drivers in my BoB, they have a separate supply
    for the output and a more sophisticated thermal/over current/over voltage management than the 26LS31, although they cost me $11.00NZD each...which is a little more than I'm used to paying
    for quad line drivers!

    I'm still not sure how to get the quadrature encoder output from the drive into Mach. Another converter?
    Who cares? What is Mach going to do with that information any way? Mach4 and the ESS are not a feedback controller solution...even if you feedback an encoder to Mach4 whats it going to do
    with it?

    Craig

  5. #5
    Join Date
    Jun 2010
    Posts
    116

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Craig, thanks for the suggestion for the AEIC7272, I'll look into it.
    As for the encoder, I assumed that ESS/Mach used the feedback for tapping, but I guess the driver must do that for it.

    Mike

  6. #6
    Join Date
    Nov 2013
    Posts
    4282

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Hi,
    Mach4 and the ESS DO NOT SUPPORT rigid tapping.

    In fact its not even Mach4 so much, but rigid tapping requires realtime synchronisation between the spindle and the Z axis. The ESS does not do that.

    If this were for a lathe spindle than the ESS would require at the least an index signal but preferably and encoder signal from the spindle to synchronise single point lathe threading.....
    but that assumes a PWM driven spindle...and your not doing that any way.

    I use Mach4 and an ESS and have a servo driven spindle, that I can use as a C axis. For example:

    g1 z-10 c3600 f36000

    will cause the Z axis to descend while the C axis rotates 10 times (3600 degrees) with the driven speed being 36000 degrees per minute, or the move will take 6 seconds., ie a 1mm pitch thread. And:

    g1 z0 c0 f36000

    will cause the tap to back out of the hole by 10mm while the C axis unwinds ten times at the same speed.

    Note that this is just plain Gcode....it does not require any special support from Mach4 or the ESS, it just coordinates the movement between two axes. It certainly coordinates between X and Y axes,
    so why would it not coordinate between the C and Z axes? It does not require the servo encoder be hooked up to the ESS.

    In this circumstance it is assumed that if the C axis is commanded to move say 3600 degrees that it has the torque authority to do so, and at the commanded speed. If it lags behind the thread will
    be inaccurate or fail altogether and break the tap. This is where IF you had realtime synchronisation between the C axis and the Z axis the Z axis would slow down a little to accommodate that the C axis
    being a little slow. This is what most understand to be what rigid tapping is and means, but it can be achieved without all that extra complication, but requires a spindle with genuine torque authority.

    I have used my 6.1Nm (cont) 18Nm (overload) C axis to do up to 12mm in aluminum (60% engagement) and 8mm (55% engagement) in mild steel. Making 100% sure that your spindle has the torque to cut
    the thread in the material at the current engagement is critical. I broke quite a few taps until I got it sorted....but it does work well within those limits.

    Craig

  7. #7
    Join Date
    Aug 2005
    Posts
    1091

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Hi,
    You can do semi rigid tapping in Mach3 with a compression/tension tap holder such as these.

    https://www.ebay.com.au/itm/16507282...mis&media=COPY

    I’ve used them successfully for a while now. My spindle motor is a 750W BLDC Motor.
    The spindle speed is controlled via the 0-10v analog control voltage output from a Geckodrive G540 into the VFD.
    You need to ensure that the spindle speed is accurate within a couple of percent. I usually tat around 500rpm. Here is an example of tapping M5 holes.

    https://www.instagram.com/tv/CZ0FzYN...d=YmMyMTA2M2Y=

    Cheers


    Peter


    Sent from my iPhone using Tapatalk Pro
    -------------------------------------------------
    Homann Designs - http://www.homanndesigns.com/store

  8. #8
    Join Date
    Jun 2010
    Posts
    116

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Craig, Thanks for the info. I didn't realize the ESS was this limited, thanks for enlightening me. I think I'm going to need to look for a different motion control that will work with the rest of my hardware that is capable of rigid tapping and spindle orientation for an ATC. I'm sure I could make it work, but I think I'd rather do it right from the start. Do you have a control you would suggest? What about this one: https://www.automationtechnologiesin...ler-industrial ?

    Peter, thanks for the suggestion. Worst case I will probably end up doing that.

    Mike

  9. #9
    Join Date
    Nov 2013
    Posts
    4282

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Hi,
    I'm not sure I agree with that characterization.

    Mach, either 3 or 4 is not a feedback control solution, and never has been. This is because Windows is not in any practical sense a real-time computing
    platform and thus any motion control must use a motion buffer. The delay imposed by the buffer prevents feedback control of servos and motors.
    This has been the case for the twenty or so years of the existence of Mach.

    This is because of Windows, so Machs competitors like UCCNC and Planet CNC are similarly challenged, ie they require a motion buffer.

    There are a few motion controllers like the HiCon, CSMIO/A and Galil that while , if you like , be 'supervised' by Mach can close a servo loop. In all cases the servo is required to be a +10/-10 VDC
    analogue servo. The motion control board is not subject to the delays imposed by the motion buffer and can therefore do real-time signal processing, is be feedback capable.

    If you require rigid tapping, in the more traditional sense of the phrase, ie the the Z axis servo be precisely synchronised to the spindle, then that is somewhat specialised. To my knowledge neither
    the Hicon nor the CSMIO/A can do that. The Galil is highly programmable and most certainly capable of doing so. Having said that I don't tink it would be easy.....but possible. A four axis Gallil
    is going to be in the region of $3000, they are exceptionally good, but not cheap.

    The PoKeys you linked to is a Step/Direction controller, and is not feedback capable, and therefore I see no strategy that will allow it to rigid tap (in the traditional sense).

    The other possibility is LinuxCNC. LinuxCNC is a real-time computing platform and thus the computer can 'close the loop' and no doubt could be programmed to execute
    synchronization of the C and Z axes.

    Any way, what's wrong with the strategy I proposed? ie using the Gcode synchronization of the C axis and Z axis? It does not require any special features at all. If you are concerned about
    the spindle having the torque authority....that is pretty much always the case with ANY tapping regime.

    Consider Mach when single point lathe threading. The motion board requires at least an index signal from the spindle, one; to synchronize the start of the thread but two; to calculate
    the speed of the spindle and thus calculate the rate at which the Z axis must advance to produce the correct pitch. If the spindle slows at all as a result of cutting then Mach can adjust
    the rate of advance of the Z axis to maintain the correct pitch. The problem is that if the speed slows by more than a few percent Mach can't keep up and the thread fails. This in turn means
    that the spindle MUST have torque authority that the speed remain constant or nearly so. Is that not exactly the same requirement I have outline with the C axis in rigid tapping?

    Lets assume you have a machine that has traditional rigid tapping, ie the motion of the Z axis is slaved to the rotation of the spindle. Lets also assume that you are tapping say a 16mm
    thread in steel with 60% engagement. The typical torque requirement for that would be 20Nm. Imagine that your spindle is limited to 10Nm. Thus when the tap enters the work the spindle will
    stall. Given that the Z axis advance is synchronized to the C axis rotation the Z axis will cease to descend also as a result of the stall. No problem.

    Imagine the same machine but that it relies on Gcode synchronization of the C axis and Z axis. When the tap enters the work the spindle will stall. It might take 50ms or so for the servo to fault 'following error'
    and signal to Mach that the spindle has stalled. In the meantime the Z axis is going to try to execute its planned move, ie descend at a given rate. For 50ms or so it will do so but the spindle has
    stalled so the tap will not descend into the hole. There could well be enough thrust or force built up over that brief period that damage to the tap or the part could ocurr. This is the nature of the warning
    I posted earlier.

    Provided the spindle has adequate torque authority the problem does not arise.

    I was going to post a video of my spindle doing some rigid tapping, but yesterday when I went to install my spindle I found that its developed a 'Illegal Hall State' fault and I can't get it to work.
    I last used it about a fortnight ago. I'm beginning to wonder if coolant has not got into the encoder in the servo and upset the Hall effect switches in there. Either way I'm off to work soon to try and sort it out.
    I've been using this spindle on and off for about six years....I sure hope it has not developed a terminal fault.

    Craig

  10. #10
    Join Date
    Nov 2013
    Posts
    4282

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Hi,
    *&%###)##!!! I have removed the encoder only to find that coolant has managed to get in and it has dissolved some of the mirror coating on the glass encoder disc.
    I wrecked it by allowing it to sit in a manner that allowed coolant ingress......my fault.

    Craig

  11. #11
    Join Date
    Aug 2005
    Posts
    1091

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Have you fixed it yet?


    Sent from my iPhone using Tapatalk Pro
    -------------------------------------------------
    Homann Designs - http://www.homanndesigns.com/store

  12. #12
    Join Date
    Nov 2013
    Posts
    4282

    Re: ESS, Mach4, SGZH Servo Spindle Motor

    Hi Peter,
    no, I have not fixed it. The encoder is buggered...my fault. I have a line on a second hand servo from the UK that matches my drive...so I'll get that in due course, 200GBP.
    I can even buy a new one but over $3000USD!!!

    Pretty annoyed at myself for letting my servo degrade in this manner.

    If I cannot repair or replace this servo cost effectively I'll probably buy a new 2kW Delta servo, 6.37Nm(cont), 19Nm (overload) @ 3000rpm (rated) and 5000rpm(max) for $895USD.

    I have another servo I bought second hand 3kW 12Nm(cont) 48Nm(overload) @4000rpm some years ago. It has a resolver...which dates it to about the mid nineties.
    I can't find a servo drive at reasonable cost so I always intended to make my own. I have almost all the bits, I have programmed a Texas Instruments TMS320F28069M up to run a
    Field Oriented Control algorithm....so its really just a matter of completing it. I'd really like to have something like a BT30 or BT40 tool changer set up.

    In the last few months I have finished my servo driven fourth axis, and have the majority of the parts for a fifth axis. The two combined represent over $5000NZD worth of parts and
    materials. I have maybe another 40 hours work, and a few hundred dollars to go to finish the trunnion table and fifth axis. I need to complete it.

    I use my mill daily particularly for making PCB's by isolation routing. It forms a big part of my business income. At this time I use a Mectronics Gmbh 800W 24000rpm spindle, and
    I use it a lot and have done for years. A smart man would not worry about my servo driven spindle which I use infrequently or a fourth and fifth axis which will likewise see
    little or at best moderate use....and concentrate on replacing my high speed spindle....it is after all the most important of the lot.

    'I want a fourth and fifth axis, and I want a high torque low speed spindle, but what I need is a modest tool changing 24000rpm spindle to replace what I have'
    The problem I face is that my decisions have been governed by 'what I want' and not 'what I need'. Which really brings into question whether I should repair the bloody thing at all!!:tired:

    Craig

Similar Threads

  1. New DIY CNC with MACH4 ESS/MB3 and clearpath servo motors
    By darko9000 in forum DIY CNC Router Table Machines
    Replies: 11
    Last Post: 07-13-2022, 03:49 PM
  2. PCNC 1100 spindle control with Mach4 & ESS
    By andysims in forum Tormach Personal CNC Mill
    Replies: 9
    Last Post: 12-24-2019, 05:23 AM
  3. Mach4+ESS or Acorn?
    By HoY in forum DIY CNC Router Table Machines
    Replies: 24
    Last Post: 11-12-2019, 04:34 PM
  4. Mach 3 to Mach4 and ESS
    By Plasron in forum Controller & Computer Solutions
    Replies: 3
    Last Post: 08-30-2019, 03:16 PM
  5. No motor movement Warp9 ESS+C11+Mach4
    By Georgie in forum Mach 4
    Replies: 2
    Last Post: 08-16-2019, 12:03 PM

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
  •