Sieg SX2PG milling machine cnc conversion

[Mark Salzedo 1Participant @marksalzedo1]

Hello,

I’m looking for the ME or MEW series starting 17 Aug 2018 for the Sieg SX2PG milling machine cnc conversion. If anyone has a copy/copies, I would be grateful to hear from them.

Thanks,

Mark

[noel shelleyParticipant @noelshelley55608]

Hi Mark, If it’s in ME then I may be able to help. Give me lssue numbers Etc.  Noel.

[JasonBModerator @jasonb]

This thread may be of use https://www.model-engineer.co.uk/forums/topic/cncing-a-sx2p-mill/

A digital subscription will get you what you want.

[Russell EberhardtParticipant @russelleberhardt48058]

I did a CNC conversion of an SX2P mill about 13 or 14 years ago.  You might be interested in by build thread done at the time:

https://www.mycncuk.com/threads/4524-Sieg-SX2-Plus-build?highlight=russell

Hope that helps.

Russell

 

[John HinkleyParticipant @johnhinkley26699]

I have been considering the conversion of my mini mil (Amadeal XJ-12) to cnc and have been reading up as much as I can, including the various posts on this site and others.  (One of) the main problems with a small machine is the lack of real estate for the installation of the z-axis motor.  I have considered two practical solutions as far as I see it.

1) use one stepper motor driving two ball screws, one either side of the spindle block/column assembly, via a toothed belt arrangement and incorporating a tensioning jockey pulley, or,

2) use two stepper motors directly connected to two ball screws, mounted as in (1), each driver being driven by a single signal from the controller.

I believe this installation would allow the use of smaller (and therefore lower torque) stepper motors albeit at slightly higher cost as well as eliminating the potential inaccuracies induced by the fairy long belt drive.

Anybody done a similar conversion?

John

 

[JasonBModerator @jasonb]

These days the other option may be to do away with the column and head. Then replace with a fabricated column, fit Linear Guides to that and a spindle motor with integral ER spindle nose on the front and a centrally mounted ballscrew etc on the rear. Or go the whoel hog and machine the X&Y dovetails off and for the linear guides there too, spaced up a bit would give a bit more room for the ballnuts.

The column could be from hollow section steel filled with epoxy & aggregate. Steel plates to mount it to the base casting. Having then got rid of the pivot point that the column tilted on you can run the ballscrew out the back and put the stepper motor there rather than get in the way at the front,

The milling machine motors are a bit slow to take advantage of relatively cheap carbide cutters and you don’t want the play of a redundant quill to contend with on some mill based machines. Though if a brushless machine you may be able to change belt ratios and get the spindle upto say 5000rpm and simply keep the max tool size down.

[John HinkleyParticipant @johnhinkley26699]

That sounds like a bit more effort than I have time or facilities for, Jason, but thanks for the ideas.  I have a high speed spindle with a built-in ER11 chuck which is really too big for the lathe, so I might try offering it up to the mill to see if it gives me any more room to play with.

John

 

[Russell EberhardtParticipant @russelleberhardt48058]

John,

Could you, perhaps remove the gas strut and replace it with a ball screw with the motor mounted off the rear of the column?

Russell

[John HinkleyParticipant @johnhinkley26699]

Russell,

Yes, that’s another angle of attack, but I was trying to get away from the use of toothed belts andpullies because of the implied backlash involved.  I would prefer to drive the ball screws directly from the stepper motor.  I have designed a mount for the X-axis, which I have printed out in ABS-GF filament, and this is a screenshot from Alibre Atom:

If I use it, I would investigate having it 3D printed in Aluminium, having been impressed with the quality of the results displayed in the thread on this subject.

Perhaps I should add that I’ve also considered throwing even more money at it and ordering a Makera Carvera cnc mill, and just keep the XJ-12 for “everyday” jobs! It’s only money, you can’t take it with you, and I have a willing beneficiary to leave it to.

John

 

[BazyleParticipant @bazyle]

You might think about what you are going to make with it initially and how you will generate the tool paths. It seems that most paths end up as going round and round at one height, step down a bit and go round again, repeat. Not so much simultaneous xyz movement. So initially you could have the programme stop at every z change and do that manually. Sure you can’t do thread milling but for a beginner in CNC is that essential?
Likewise if the conversion to ball screws is daunting a little work to eliminate backlash allows the existing leadscrews to work and give you a feel for whether the use you put it to justifies the effort.

[John HaineParticipant @johnhaine32865]

If you choose the right belts and pulleys there will be no “implied backlash”.  My Denford Novamill has belt drive on all 3 axes and backlash is minimal and mainly due to the nuts – ~0.01mm.  This is compensated in Mach 3.

Personally I think there is no point in not having Z axis drive – you will inevitably have to use CAM to generate g-code sooner or later and starting by hand coding to avoid Z moves will be error prone.  Even using wizards will demand Z moves.

I also think that ballscrews are a must.  Trying to adjust the backlash out of ordinary screws will be pretty difficult and probably increase friction at some point of travel which may stall the drive motor.  When I first converted my lathe I tried to use an ordinary screw on the cross-slide and it was not a happy experience – I fitted a ballscrew which is direct drive.  On a lathe you can get away with a normal leadscrew but need to take care with programming, but that’s not possible on a mill.  Backlash compensation in software sounds ideal but in practice making it work with anything but very small backlash is very hard. By the way a good reason for having ballscrews is so that you can climb mill without worry – software compensation can’t stop the table being pulled through the backlash if the cutter grabs.

I don’t know what you have in mind for the control software?  I would recommend that you AVOID Mach 3, and particularly using it with a parallel port.  The company who supplied it has recently been swallowed and even Mach 4 seems to have been de-prioritised; and the Mach support forum has moved too.  Mach 3 has not been maintained for a decade or so and manages to keep going with a variety of user kluges – sooner or later Microsoft will introduce a Windows update that breaks it that no one will fix.  Mach 4 also seems to be a mixed bag.  I do use Mach 3 mainly through inertia but I think quite soon I will have to ditch it.  I would probably move to UCCNC for my mill as I use their motion controllers but the lathe is a different matter.  Other options are LinuxCNC or something based on GRBL plus a suitable g-code streamer.

[John HinkleyParticipant @johnhinkley26699]

Thank you for further input, gentlemen.  I am not a complete novice at cnc machining, however. After I was given a Shapeoko router as a competition prize, I started using Fusion 360 for the production of gcode, with the use of the built-in post processor. Quite successfully, if I say so myself, at least to my satisfaction.  Made quite a lot of mdf dust in the process but the gantry axis arrangement I found somewhat lacking in rigidity. The results of the early trials can be seen on my YouTube channel, Bizzi Bee Workshop.

Should I decide on the self build route, I will certainly revisit the pulley/belt drive, after the reassurance that John Haine has given in his response.

John

 

[JasonBModerator @jasonb]

I agree with John, if you are going to do it then do all three Axis.

Although the cutter might go round and round the outside of a part what do you do if you want that part held by tabs where the cutter needs to lift and drop at each tab

What do you do if there is a pocket and the tool needs to ramp or helical cut down into the pocket before it starts going round and round.

What if say the part is “H” shaped, do you waste time going round the outside when cutting the material out top and bottom or do you want the tool to lift and make a rapid move from one area to another.

And all of that before any 3D work is done which at least for me I do a lot of. Even the first trials I did with the KX3 still on the pallet were multiple height items with tabs and ramp downs.

No Z also means you are going to spend a lot of time stood infront of the machine rather than being able to walk away and get on with something else while the CNC gets the job done.

[JasonB]

On JasonB Said:

…..

No Z also means you are going to spend a lot of time stood infront of the machine rather than being able to walk away and get on with something else while the CNC gets the job done.

Not to mention lots of time spent hand coding or modifying CAM generated code.  And, at least with Mach 3, pausing program flow and restarting seems to be iffy to say the least.

[BazyleParticipant @bazyle]

Good grief. I didn’t mean you should not have Z control for the next twenty years, just to get you going while you decided how you were going to fit the Z screw which the earlier writer has said was a problem. You might then be able to use the embryo cnc to make the parts you needed for the next stage.

[Mark Salzedo 1Participant @marksalzedo1]

Thanks for all the information. Noel I’m trying to get the issue number. If so, I’ll post it on the forum here. Thanks, Mark

[JasonBModerator @jasonb]

Issue No is in the link I posted in my first reply, 11 parts in alternate mags.

[Michael GilliganParticipant @michaelgilligan61133]

For convenient reference:

.

.

MichaelG.

[John HinkleyParticipant @johnhinkley26699]

Just a quick update on my situation with regard to this.  This morning I placed an order for a Carvera Air mill with 4th axis and Laser accessories. As part of their Spring sale, the package includes a pack of milling cutters.  In due course, I hope to obtain 4mm and 6mm collets and more cutters to suit.  I’ll start a new thread when it arrives.  That won’t be very soon, what with it coming by surface transport, I understand, not to mention the uncertainty of the Middle East situation.

John

 

[JasonBModerator @jasonb]

I look forward to the thread about how you get on and what it is like cutting metal. Videos I have seen look promising.

[John HinkleyParticipant @johnhinkley26699]

You and me both, Jason. I have yet to decide what my first project will be.  I am reluctant to follow the herd and churn out one of the “packaged” ones that come with the machine as a starting point. Good job I’ve got a while to come up with something! I’ll be designing it in Alibre Atom and producing the gcode with Fusion 360, having used that route for the Shapeoko router previously with reasonable success (eventually).

John

[Engine BuilderParticipant @enginebuilder]

I have the Makera Carvera Air and am very pleased with it.

Their included Cam software is very easy to use.

If anyone wants the laser attachment and pc making kit let me know, I don’t use it.

[BazyleParticipant @bazyle]

I hope they are better at producing the machine software than they are at website software. I tried to find out a bit about the product from their website and found it a one of the worst I have ever encountered. Desperate to avoid providing real information it also keeps resizing pages to small. Gave up when the text pages shrank to 6 point on my laptop.

[Engine BuilderParticipant @enginebuilder]

Bazyle, The website works fine for me on my desktop.

[John HinkleyParticipant @johnhinkley26699]

Works fine for me, too, on iPad and desktop. Using Firefox browser on both. Can’t say about a smart phone as I don’t use one for browsing, just making and receiving calls.

John

 

[Author]

Posts