Page 1 of 3 123 LastLast
Results 1 to 15 of 31
  1. #1
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default Arboga Manual to CNC Conversion

    I've completed the first stage of my manual to CNC Arboga mill conversion, I decided to use a ball screw hung off the side, a 700N gas spring is mounted on the other side to balance/reduce the weight, IF - and that is a big if - this works OK I can lock the quill in place for CNCing and still have the use of the quill if I need to. The other option was utilise the existing rack and pinion, this would have relied somewhat on gravity to eliminate any backlash - a ballscrew seems to be a more positive system and any backlash would be from the ballscrew and nut plus any flex in mounting system/drive belt.


    I initially mounted just the gas spring on it's own and found that it more than balanced the weight and had a slight positive affect on the system - once the stepper and ballscrew (which are not light) were fitted I estimate the overall weight of the Z axis system that the stepper has to lift is around 5-10 Kg plus any friction, the gas spring unfortunately does not reduce the mass so the stepper still has to contend inertia plus the weight + friction.


    The ballscrew is 25mm x 5mm pitch the drive belt is 18mm wide 5mm pitch with a 2.5:1 ratio - the stepper is closed loop Nema34 12Nm 1.8deg per step, these are not a full closed loop system but a sort of hybrid system in that the stepper provides feedback to the stepper driver so the driver can try to regain lost steps should these occur, if the lost steps can not be made up the driver can alert/signal the motion controller - how the motion controller handles this depends on the motion control software and is a bit of an unknown for me at present. The 2.5:1 ratio multiplies the motor torque - it can easily be changed with different pulley sets and possibly removed altogether if I mount the motor directly above the ballscrew for direct drive.


    The ballscrew nut mount is a section of 70mm square tube that was left over from a long ago carport build, the ballscrew top bearing and motor mounts are made up of 100mm wide Al plate, 15mm thick for the bearing mount and 10mm thick for the rest.


    With everything bolted up tight and a DTI placed under the quill I could not detect any backlash when I rocked the drive pulley back and forth by hand, next step for me is to temporarily wire up the motion controller, bumble my way through the setup procedure and do some tests on backlash and possibly speed tests as well.




    This all new to me, while I think I'm headed in the right direction with what I'm doing, I may well not be - I did vow at one time that I would never venture into 3D printing or the CNC world - I've had my 3D printer for over a year and now this.

    20210727_135439.jpg 20210727_135553.jpg 20210727_135630.jpg 20210727_135635.jpg

  2. #2
    Join Date
    Nov 2017
    Location
    Geelong, Australia
    Age
    57
    Posts
    2,647

    Default

    Good progress!
    I think you’ve said previously but I can’t recall the answer - what controller/software are you going to be using?

    Steve

  3. #3
    Join Date
    Jul 2016
    Location
    Melbourne
    Age
    35
    Posts
    1,519

    Default

    How will you deal with the column locks? I have a very similar modig drill and its not much of a mill but its a great drill.

    Sent from my SM-G973F using Tapatalk

  4. #4
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    Controller is DDCS Expert stand alone(offline) 3/4/5 axis I think it is the same as a FOINC 350 - reading posts from various forums most use the Fanuc post processor, there is an active Facebook group with what looks like a lot of users but not a lot else so if I have issues to solve I'll be pretty much on my own - I did a lot of looking around and did not really find any 'turnkey' solutions for manual to CNC conversion unless you go the route of buying a fully built and working CNC mill - my budget does not extend that far.


    The column locks - I have been using it for quite a while without locking the column, took a while to get the column gibs just right - a good cutter sharp cutter and easy does it, I have never driven the mill hard - locking the column helps if using a single point cutter like a boring head or a flycutter, I can't see myself using them in a CNC situation. As I said it is all new to me - using the column for the Z axis may turn out to be a bad idea, if so I can always do what many people have done and use the quill - a stepper on the fine downfeed, I've purchased a small stepper for the 4th axis I have planned and can always use it on the quill, another thing that I could possibly try is a solenoid on the column lock, the controller has 24 outputs for driving such things as coolant pumps, air lines etc it may be possible to operate a solenoid from one of the outputs.

    Screenshot 2021-07-27 223518.jpg Screenshot 2021-07-27 222814.jpg Screenshot 2021-07-27 222435.jpg

  5. #5
    Join Date
    Oct 2011
    Location
    Norwood-ish, Adelaide
    Age
    59
    Posts
    6,538

    Default

    Quote Originally Posted by familyguy View Post
    ...using the column for the Z axis may turn out to be a bad idea, if so I can always do what many people have done and use the quill - a stepper on the fine downfeed...
    As a general rule, keeping the quill as short as possible while milling reduces wear and tear on the mill. Driving the table up and down will give you a more rigid set up.

    Quote Originally Posted by familyguy View Post
    The column locks - I have been using it for quite a while without locking the column, took a while to get the column gibs just right - a good cutter sharp cutter and easy does it... another thing that I could possibly try is a solenoid on the column lock, the controller has 24 outputs for driving such things as coolant pumps, air lines etc it may be possible to operate a solenoid from one of the outputs.
    That is something that is worth looking at, as once again it will increase rigidity & improve surface finish.

    Michael

  6. #6
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    I cobbled together some components and in the first instance powered up just the motion controller to set up the Z axis parameters. The large square transformer is 245v to 215v - this powers the 2 small 24Vdc power supplies - 1 for the motion controller system power and the other for the IO power that provides voltage for control signals to the stepper drivers and anything else that may be connected to either the input or output ie coolant pump relay, limit/proximity switches etc. In theory I shouldn't need this type of drop down transformer - but the 24Vdc power supplies are 220Vac input and my mains voltage hovers around 240V to 247V so rather than take chances I'll be using this transformer in the final build.


    The toroid transformer is one I had earmarked for a power amp project I intend to build in the future - it has 2x50V windings so for the test I'll be using 1 winding for 50VAC into the stepper controller - the stepper controller can take either AC up to 80V or DC up to 100V.
    After double checking connections and voltages etc I power it up and there was panic for a few seconds when the screen did not come on instantly but once the loading system message appeared I was able to start breathing again, the screen is colourful, has a lot of pages/screens of information and I can see that for some time I'll be going around in circles wondering how to get back to a particular menu or screen, I eventually found my way to the axis setup screen but didn't get very far, I found I couldn't change any parameters unless I was signed in as a supervisor - time to go back and read the user manual.

    20210728_150022.jpg 20210728_150409.jpg 20210728_150716.jpg

  7. #7
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    After sorting out the password, there are 3 levels of access (bottom level is 666666 - double devils number, is this an omen?) I still could not get any movement from the Z axis, - it dawned on me eventually that the stepper driver dip switches may not be set and sure enough they were set for maximum micro-step, with this reset for no micro-stepping I had up and down jogging. The motion controller user manual is not bad and does not read like a google translation from Chinese to English - but even so I could not find where to jog a set number of steps so I could do some preliminary checking of backlash so I turned to the MPG controller. Plugging this in gave me nothing, checking and changing some of the parameters for this device still gave nothing, not knowing if it had some sort of fault like loose wires or bad solder joints or if there were some other settings deeper in the motion controller that I hadn't found I decided to open up the MPG controller and have a look, while the build quality is ok I saw a few wires that had been extended, not great but what do you do when the wire has been cut off too short to reach the solder joint, I did find one bad joint, the wire had come away completely, after fixing this didn't change a thing, the motion controller has a screen that shows the status of all of IO ports - MPG port is one of these, I could see the rotary encoder was working but there was no other activity - something is not connecting with the IO power supply, then it hit me I could see I had not connected the 24Vdc supply at all, great work I thought. With power to the IO ports, the MPG (manual pulse generator) worked seemingly as it should, I was able to do a preliminary test for back lash - DTI under the chuck and zeroed, move Z axis up one full turn of the rotary encoder up and then back down one full turn, the DTI returned to zero, so far so good but I feel there must be more to it as far as checking for back lash, I'll need to do some more youtubing.

  8. #8
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    I've been making slow progress with my CNC conversion - even though I vowed to do no more electronics repairs, I guess I'm just too soft and found my self with an amp, 3 speakers and 4 output transformers to wind, I did though manage to farm out the table and get it machined to allow clearance for the X axis ballscrew nut - the Z motor mounts were done and tested a while back - the X and Y motor mounts are finished and I am at the stage where I could reassemble everything and only have the electronics in the control cabinet to mount and wire up. But with the XY table on the bench now would be the time to add some lubrication grooves and external oil points to the dovetails and also the ballscrews, the mill had no lubrication grooves at all. I did a trial assembly of the Y axis and connected it up to my jury rigged controller setup, not surprisingly there is no backlash I uploaded 3 small videos to you tube, Z axis trial, Y axis trial and using the MPG controller to elongate a mounting hole on a Chinese XY table.


    While the XY table has been out of action I purchased one of the el cheapo Chinese XY tables from the 'bay', it can be seen in the video sitting on the floor - I mounted it on the mill base so I still had a functional mill that I could use in making the X and Y motor mounts - also my ballscrews did not come with keyway slots which I need for the 'flexible' couplings between the motor and ballscrew, while they are not really flexible they do tolerate a small misalignment that might creep in.


    The Chinese XY table is not great but adequate for what was asked of it, I found I had to adjust the gibs constantly for the first few hours of use and it seemed to get better the more I used it. It is obviously made as quickly and cheaply as possible and from viewing youtube videos the build quality of them is very variable, mine does not seem too bad - I did notice that not one edge had been broken so I ran over every corner with a file before the inevitable happened. If the MPG controller turns out to be PIA for manual milling then I may use the Chinese XY table bolted to the main mill table for used as a manual mill for quick jobs like cleaning up a saw cut.


    The aviation style connectors I ordered finally arrived and I realised that I did not take a great deal of notice of the size, the 6 pin connectors are ok for the stepper feedback cables but not for the main wiring that powers the motors, pins are small and I was not keen to have 6-8 amps going through them - I located some ex military connectors in Lithuania, seller was very helpful, they arrived quickly and look to be up to the task.

    20211101_211227.jpg s-l1600.jpg s-l1600 (1).jpg


    https://www.youtube.com/watch?v=QpD7WQyAAQs
    https://www.youtube.com/watch?v=p1nUxvbOujo
    https://www.youtube.com/watch?v=bDaxn1OnPUQ

  9. #9
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    I'm getting closer to finishing the XY ballscrew and stepper mounts, I decided to go with a simple gravity feed oil system for the slides and ballscrew - the slides and ball screws are fed via a small bore (4mm OD) copper pipe which leads back to an oil cup mounted on the rear ballscrew cover - one for each axis. I tested this with the XY table assembled on the workbench using ordinary 30/40 motor oil from my workshop oil can, takes about 3-4 hours for the oil cup (5ml when full) to empty, I notice that the motor oil does create some drag so when it is finally finished I'll investigate the use of something less viscous. I know it is not ideal to have the ball screw on the same feed as the slides but looking at other diy conversions on youtube I see many don't bother with oil feed to the ball screws and I figure that any lubrication is better then none, being for hobby use I can't see the ballscrews wearing out in a hurry, in addition if they wear out they are readily available - not like the original lead screws - acme thread, 0.2inch pitch, 0.875 inch diam, I'm expecting to add 2-3ml oil every few months.


    The X axis ball screw is covered by the table but the Y axis ball screw is partially exposed so I made front and rear covers/shields for it, the covers are wide and deep enough to cover the sides as well, these were made from 2mm thick steel plate, my folder had no hope of bending 2mm steel plate so I cut out the individual bits and tried stick welding, the first cover warped badly and while I was able get it sort of straight it looked crap so had to be thrown out, for the next try I used the oxy and fusion welded it together - by the time I was nearing the end of the 2nd cover I was getting pretty good to the point where I didn't need to file/cleanup the welding and the best part is not even a hint of warp. The motor mounts are fabricated from heavy angle iron and 6mm plate, every thing slid together nicely with almost no issues, but - the X axis had a bad shudder in one direction only when the ballscrew was turned by hand, I eventually traced this down to the ball screw being a fraction too long and rubbing on the underside rear of the table, just touching it, I'll cut a few mm off the end.


    The clearance machined for the Xaxis ballscrew mount broke through to the top of the table so I had to make up an cover, held in place with 4x3mm screws, once it is up and running I'll seal this with a bit of silicon - just in case I use flood coolant at any stage.


    I still have a way to go - I've wound 2 transformers so I can use linear power supplies, a single transformer for the Z axis and a larger toroid with 2 secondary windings for X and Y axis, I was going to run straight AC to the stepper drivers but am now leaning towards a bridge rectifier and reservoir capacitors.
    20211117_112949.jpg 20211117_115543.jpg 20211117_194828.jpg

  10. #10
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    Made some more progress on the CNC conversion, I don't really have that much more to do to make it functional but it still seems to be taking forever - after looking at a number of options for a control box I decided to incorporate it into a small cabinet with benchtop and drawer, made with 25mm squ tube steel frame and sides covered with colour bond steel sheet, a door on the side and a removable top give reasonable access to the inside. The motion controller sits in a sloping front metal enclosure that is slightly raised on a 60mmx25mm steel tube, all the cables to and from the controller run through the tube. The cabinet holds the fuses, power supplies, VFD drive and the stepper drivers. All drive and control connections from the mill to the control box are with aviation style connectors, 3Ph power from the VFD to the mill is via a standard 3ph plug and socket. I used a scrap bit of Al the for the front panel of the motion controller cabinet, once it's working I'll make a new one and do the engraving on the mill (I hope).
    20211209_210321.jpg 20211209_213122.jpg 20211209_213129.jpg

  11. #11
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    The X and Y axes are all wired tested and working in jog mode - parameters set for the micro stepping setup I'm starting off with - not sure about micro stepping I understand that it increases precision at the expense of speed - I have started with 800 steps/rev against the motor default of 200 steps/rev this should give me around 2 tenths of a thou resolution. The motors are connected to the control cabinet via aviation style connectors a large 4pin for the motor drive and a smaller 6 pin for the feedback encoder, I didn't really pay too much attention to the size of the small connectors when I ordered them and found they are quite small, even though I've been soldering for over 1/2 a century I did find them quite fiddly and silently cursed myself a few times while soldering.
    I thought I'd better come to grips with something simple to try as a first project - engraving a new front panel for the motion controller box - I could see that the mill would be operational soon and I wouldn't know how to use it - I bumbled my way through Fusion360 one evening to come up with a small engraving NC test file using a Fanuc post processor. From experience I know that it is asking for trouble operating machinery if everything is not properly connected and secured but I couldn't resist trying it out in 2D - felt pen in place of a cutter - the Z axis wiring is not complete/working so the complete tool path is drawn out - link to the video - not very exciting stuff but gratifying for me to see.


    https://youtu.be/gXUPXkv5Ql0

  12. #12
    Join Date
    Sep 2009
    Location
    Newcastle
    Posts
    341

    Default

    Microstepping improves smoothness (reduces vibration) more than anything. Cannot be relied upon for accuracy.

    https://hackaday.com/2016/08/29/how-...epping-really/

  13. #13
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    Before all work stopped for Xmas and the endless meals etc. I managed to make some progress on the conversion and all 3 axis now working - even though I was keen to do some trial cuts, I was a bit hesitant - a manual machine gives you some feed back through the lead screw handles that you don't get with cnc, but at the end of the day the worst I can do I thougfh is break a cutter. I ran into an issue and while I can work around it it would be nice to sort it out, at this stage I'm not sure if it is a Fusion360 thing that can be fixed with a check box or if it is an issue with the motion controller setup. In Fusion I set the lower left corner of the work piece as the ref point or origin when setting the tool path and generating the nc file - on the mill however the part will be mirrored so I have to indicate the top right, which effectively makes the part upside down if I stand at the back of the mill then it looks right - I have some research to do here.


    For the test cut I machined a 3mmx3mm groove around the edge of a piece Al from a failed project, the need for the workaround aside it was pleasing to see it worked - could I have done this manually - clearly for something as simple as this the answer is yes and it would have been quicker. A link to a video - not very exciting plus a bit shaky I was trying to keep the Estop button in sight, just in case.


    https://youtu.be/IgIwLlxmeOA


    I still have a fair amount of work to do - tidy up the stepper motor cables, currently just hanging on the ground, organise a coolant/air blast system, make a top and drawer for the control cabinet , hang the control cabinet door and make and engrave a front panel for the motion control enclosure.

  14. #14
    Join Date
    Sep 2012
    Location
    York, North Yorkshire UK
    Posts
    6,414

    Default

    Hi Familyguy,

    Check the rotation direction of the X - Y steppers ! If one is running in the wrong direction that is the effect you get ! A bit like swapping phases on a three phase motor.
    Best Regards:
    Baron J.

  15. #15
    Join Date
    Dec 2007
    Location
    Adelaide
    Posts
    571

    Default

    Quote Originally Posted by BaronJ View Post
    Hi Familyguy,

    Check the rotation direction of the X - Y steppers ! If one is running in the wrong direction that is the effect you get ! A bit like swapping phases on a three phase motor.
    That seems logical, the direction of the steppers is easily changed in the stepper driver and in the motion controller parameters - however the motion controller has a set of jog buttons, changing the direction of the motors would reverse the jog button operation, pressing left jog button would move the table to the right and I feel that's not intuitive and asking for trouble to operate this way. At present moving the table to the right is shown as a positive direction (counting up) on the DRO, and this at first glance this seems right but as the Z axis is stationary the Z axis movement is in a negative direction (to the left) relative to the table - the DRO should be counting down. I'm hoping that one of the 250 setup parameters will fix the issue, some of the parameter descriptions in the manual are a bit confusing and I'll need to set a few hours aside to play with the settings, probably after Xmas - the next 5 days aren't mine but have been given to endless road trips, meals and cooking.

Page 1 of 3 123 LastLast

Similar Threads

  1. New manual to CNC conversion question
    By familyguy in forum CNC Forum
    Replies: 8
    Last Post: 15th Jun 2021, 05:53 PM
  2. Arboga E 825 followed me home yesterday :)
    By jhovel in forum METALWORK GENERAL
    Replies: 10
    Last Post: 24th Nov 2016, 11:52 PM
  3. A DRO on my Arboga
    By familyguy in forum METALWORK GENERAL
    Replies: 0
    Last Post: 20th Oct 2016, 11:29 PM
  4. Wiring VFD to Arboga mill?
    By neksmerj in forum METALWORK GENERAL
    Replies: 46
    Last Post: 15th Mar 2011, 02:25 PM
  5. Arboga
    By Anorak Bob in forum METALWORK GENERAL
    Replies: 7
    Last Post: 15th Feb 2011, 12:00 AM

Posting Permissions

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