[proengines]

Hi all, I've been a member here a while but haven't posted a lot. I thought someone may be interested in this. I picked up this lathe a few months back with the intention of doing this to use mainly for gunsmithing work. I already have a Colchester Master 2500 lathe but didn't want to rip it apart to do this just in case I ruined it and this little Hafco has a very short headstock so is perfect for barrel work. This lathe came from a local university workshop and had done very little work so it was a good candidate.

Here's a bit of a rundown on where I've got to so far over the past few weeks. i'm copying and pasting some posts from another forum that I've posted up over that time so it may read a little strangely until I get up to where I'm currently at:

As it came, mid way through cleaning all the packing wax off it:


With all the bits removed that I wont be using any more, basically the feedscrew/leadscrews, the gearbox for them, the apron and crossfeed screw and handwheels, the compound rest etc..

[proengines]

Parts-wise, here's what I went for:

2 of these, 1200 in/oz stepper motors:


A leadshine integrated breakout board and stepper drivers.


2 48V, 100W switchmode power supplies:


An index pulse care to give a spindle speed feedback signal to the breakout board, I just need to make a matching chopper wheel and a bracket for it to mount to the lathe.


I got a 16mm ballscrew for the x and a 20mm one for the Z plus some support blocks etc..

[proengines]

Here is the start of some actual work to fit everything up:

Saddle machine out to fit ballnut:







Ballnut mount machining:













I had to make up a bracket to space out the ball screw support bearing and sit it at the right height. I though I might as well incorporate the stepper motor mount in it as well, All the machined aluminium bits are 6061 aluminium:







This is the sliding mount that the motor attaches to to allow for timing belt adjustment:







It bolts onto the larger chunk with 4 bolts through these slots to allow it to slide for adjustment:



Here it is all fitted up to the lathe, this assembly will bolt to the mount for the z-axis ball screw when I get it done to make it all a bit more solid:

[proengines]

I machined the pulleys to suit, even did some keyway broaching in the lathe. I'm still waiting for the proper belt to turn up, this is an engine timing belt that will do the job for now:



I got the z-axis ballscrew mounted:





I got the rest of the saddle parts made including the Z-axis ballnut mount. I cut it from 50mm plate to reduce any flex:







I also made a template for the chip cover which will run the full length of the lathe to save the ballscrew/nut from and chips or dust. I'll get the real one folded up from 1.6mm stainless. I cut a channel in the ballscrew mount so it could fit in (All credit to someone on this forum who I borrowed the idea from, I can't recall the name but I liked the idea):



I also got the rear support for the x-axis ballscrew made and fitted up:





Today I got the z-axis stepper motor mounted, it uses a cushioned coupling which apparently work really well. If it flexes too much I'll swap it for a solid one:



It all fits within the original rear belt/gear cover so I'll just make a new front panel that lines up with that which will give me room to mount an emergency stop and cycle start etc.. buttons.



I might have a go at some of the electronics this weekend and see if I can get it to move a little. I'm still waiting for the limit switches to arrive so I can mount them up and need to make a riser block to replace the compound slide.

[proengines]

I started on the disc for the spindle speed sensor this morning and decided that I could kill 2 birds with the one stone if I incorporated the disc as a part of the rear spider, it's looking as though it'll work ok:

[Fastest1]

Nice work. I like your approach to a few of the problems.

I would think your mounts should minimize flex. Didnt you say they are 2" thick? I have a Central Machinery (HF) 12 x 36 and have considered converting. I might have to get around to it.

Is it the photography or is the hole for the X screw off center in your mount? That is quite an elaborate mount with the plug entering the ball screws bore. The 2 bolt holes would have likely been sufficient.

Ok I wasnt going to say anything about your stepper mount for the carriage. I thought the additional plate for the stepper itself with the slots was bulky and an after thought in relation to everything else. Then on post #3 picture #14 down on the right side, there is the mount like I would have liked it. What went wrong with that mount? Not enough clearance of the stepper to the bed?

Also on the pillow block for the X, right where the pulley is. Is that just proud enough over the ways for the carriage to hit it? As in a positive stop? Otherwise over travel would have the carriage hitting the pulley. Wouldnt it?

[proengines]

Nice work. I like your approach to a few of the problems.

I would think your mounts should minimize flex. Didnt you say they are 2" thick? I have a Central Machinery (HF) 12 x 36 and have considered converting. I might have to get around to it.

Is it the photography or is the hole for the X screw off center in your mount? That is quite an elaborate mount with the plug entering the ball screws bore. The 2 bolt holes would have likely been sufficient.

Ok I wasnt going to say anything about your stepper mount for the carriage. I thought the additional plate for the stepper itself with the slots was bulky and an after thought in relation to everything else. Then on post #3 picture #14 down on the right side, there is the mount like I would have liked it. What went wrong with that mount? Not enough clearance of the stepper to the bed?

Also on the pillow block for the X, right where the pulley is. Is that just proud enough over the ways for the carriage to hit it? As in a positive stop? Otherwise over travel would have the carriage hitting the pulley. Wouldnt it?

Hi fastest,

I had a length of 50 x 100 bar sitting around so it got the nod for the mounts, the rest of the saddle is from 25mm plate for the same reason, I had a big plate of it sitting here. I figure that given I'm using aluminium for most of it rather than cast iron a bit of overkill is in order.

Yep the hole for the x screw is offset upwards to give the ballnut clearance on the bottom side, it would have been nice to have it on centre but where it's sitting now it has about .5mm clearance to the v on the lathe bed so to fit it inside the cross slide that's where it had to go. The mount locates with the plug and also the shoulder of the pocket sits against the bottom of the cross slide to stop it rotating, that way the screws don't have to provide any radial holding, they just hold the mount against the cross slide. I made it from 50mm plate to give the cross slide the extra travel in -x. I remachined the ballscrew to give a larger diameter and longer spigot for the pulley to mount to which is why the lock nut is on the outside of the pulley. The bearing housing does sit a touch proud and would stop the cross slide, I could have machined the housing to clear but it will have a limit switch that stops it before it gets to there. There will also be a cover over the pulleys and belt which will sit higher that the mount.

That first plate I made that's sitting on the bed was just to work out where to put everything and to see if my measurements were right with the ball screw height and mounting setup. I needed to mount the x motor lower to clear the z ball screw and realised when I was done that I need to move it towards the bottom of its adjustment to clear the z ball screw cover as well. The motor is mounted in the sliding plate to bring the shaft out far enough to engage enough of the shaft in the pulley when it's lined up with the top one. I'll see how that works, my other thought was a fixed motor mount with a couple of eccentric belt tensioners which could still be an option if this doesn't work.

The whole thing is a bit of a work in progress and some parts will probably change as time goes on but it's an interesting process.

[embraced]

Nice job.

Your choice of a 1200oz/in stepper is insanity to me though - you'll be getting over 10kN (1 metric tonne) of force through your axes.

[proengines]

Thanks Embraced. I had a look at a few of the commercial machines around and the microkinetics one is running around 1600 in/oz on the z axis. I just figured for ease of setup I'd run the same motors on both. Without trying to ask a silly question, where is the downside of having more power than you need?

[embraced]

The only downsides would be the risk of destroying something in the event of a crash, and lower top-end speed - the torque of those big steppers drops off rapidly.Have a look at this graph: http://www.anaheimautomation.com/ima...s(800x549).png

Other than that (which is not a problem at all) your build is excellent. Very professional job.

[proengines]

Ah ok, got you on that. This is a first for me so I'm always interested in learning some more. This lathe is really a toy to learn all of this stuff. I've got a Tormach PCNC 1100 which I love and has paid for itself many times over so I thought a lathe to compliment it would be handy. I looked at the Tormach lathe but it's not quite what I'm after, I want a conventional machine that is CNC controlled rather than a gang tooled lathe. I might make up a mount though that attached to the T-slots on the rear of the saddle so I have gang tooling capability if I do decide I need it, at worst I think I'll mount an upside down parting tool on the rear to start with.

I got the spindle speed pickup mount drawn up today but ran out of time to make it. I also mounted up the new spindle motor, it's a 1.5kw 3 phase motor which I hope will do the job nicely. I'll hook up the VFD to it and make sure it works before I go too far. I had the VFD here already, it's a Telemecanique Altivar 15 which is 3 phase in and out which I bought for an old cylinder head machine I had but never fitted. I'll run it like that if it works or if it's a hassle I'll get a 240V single phase VFD which makes the machine a little more garage friendly if I ever move out of my current workshop.

[ebrewste]

That's a really nice job. I've been considering converting my lathe in a way similar to this. It's great to see so many details thought through that I hadn't considered.

How far away are you from getting it working. It seems like you haven't started on getting a computer going, but the electronics are in place? Great progress!

[proengines]

Ebrewste, I figured I know how to make parts so I'd get that out of the way first before I delve into the dark art of electronics. Tomorrow will be the spindle speed sensor bracket and a pulley for the drive motor then it's on to the electronics. I've got a computer here to use (a Dell optiplex similar to what I have on my Tormach) as well as the stepper drive, spindle sensor, limit switches, power supplies and an enclosure to put them in. I bought some wire and bits and pieces the other day and I've pretty much run out of things to make so it's time to start wiring stuff up. I want to do a dry run first before I fit some proper connectors so I'll rip the motors back off the lathe and try it all out on the bench first.

I reckon I'll have it running in some form next week, I've got 6 days off work so it's a good opportunity to get into it. I've got a rifle stock to finish making first but I think I only have a day left in that so I can concentrate on this.

[proengines]

I finally had a go at some electronics and after a few false starts it's got some motion!

Cnc lathe conversion first moves - YouTube

I spent ages getting a computer sorted that would run Mach3 properly and while I was out chasing bits,came across these tiny solid state computers, 1Ghz processor and enough ram, they run flash memory for a hard drive and it was tiny but a USB stick plugged into works fine as a hard drive and it passed the driver test that Mach3 provides which is weird given it has onboard everything. The best part is it cost $50 brand new as they were leftovers from a job that apparently didn't work out. I've got a couple more for an upcoming router project and a spare. If they end up not working properly it's not a lot of money down the drain I suppose.

I did a basic mathematical calibration of the motors and it seems pretty close, the digital calipers held on the ways read 50mm when I moved the Z 50mm. I'll get a dial gauge mounted up to check it exactly when the time comes.



You can see how small these things are with the Bic lighter sitting on top, it'll fit right inside the control box which is a bonus.



Here's a screenshot of the control screen, I used the MX-3660 XML file and did a bit of editing to use the lathe screen. This is all new to me but seems to work ok.





This is the integrated breakout board and driver module, no inputs connected yet, just the basics to get it moving. It saves a lot of fiddling around with wiring which I'm no expert at so to me it's a good thing.



Bad news if the VFD I had here for th spindle motor doesn't appear to work so it looks like I'll be ordering a new one this week. The bonus I suppose is that I can go single phase now which is what I was really after.

[proengines]

I had the variable frequenct drive arrive for the spindle motor last week and got into it a little this weekend.

I got the spindle speed sensor mounted, it straddles the chopper wheel on a little aluminium mount I knocked up. The good news is that Mach3 finally reads the spindle speed after a lot of playing around and editing, there was most of a day taken getting to that point.





The VFD is a lot smaller than the old one I had, after a quick google apprenticeship in electrics I got it wired and it ran the motor straight up. I can control the motor speed directly from the VFD but still have a little work to to getting it running closed loop from the controller. It's only a setting here or there I hope.



Once I got to this point, I thought I'd try everything else out before mounting everything in the control box. Of course knowing my luck, the steppers on the axis had stopped working sometime throughout my fiddling to get the spindle motor running. About 4 hours later after a re-install of Mach 3, removing and replacing drivers, rewiring the limit switches and lots of other attempts to get them working I noticed I had bumped the charge pump jumper switch. I fixed that up and everything was in motion again.

I also got the z-axis limit switches properly mounted as well:





The X-axis switch is a little trickier, I got a sealed aluminium box to mount it in today so it's next on the list. Once I knew everything worked I unplugged and unwired everything and started to mount it all up in the control box enclosure, here are the 2 48V power supplies for the steppers, the VFD, the stepper control unit, the spindle speed sensor board (with the dodgy looking 5V power supply I hacked up from a Jaycar wall wart transformer). All I need to do now is route some power neatly to everything and fit some fuse blocks on the delicate bits then mount some connectors for all the cables to plug into the bottom of the box.



The control unit, spindle speed sensor board and its power supply:



In the process of trying to get everything to work, I set up the shuttle controller from my Tormach which worked great, I'll be shopping on eBay for an extra one tonight.

It's back to making some parts for me this week but I might squeeze a little more in on the lathe during the week.
.

[RotarySMP]

Nice work. nice clean design.

How did you decide where to locate your Z end switches? I find I am constantly swapping between collets, three jaw, four jaw and face plate set-ups and geave up on a Z end stop as it was never in the right place to add any safety.

You can probably crank that acceleration way up from there, it sounds kind of slow in the video.

[proengines]

Thanks Mark and good question which I don't have a good answer for. The switches currently stop the saddle crashing into the ballscrew mounts which I figured was a good start for a worst case scenario. I suppose I can set the soft limits to take care of anything else if I need to but I can understand how the difference between different chucking systems make it a little redundant.

The accelleration was set very slow in the video as well as the rapid speeds, I just wanted to see if it worked, I had a play with the motor tuning yesterday and it can move scary-fast compared to my Tormach. Once it's properly up and running I suppose I'll find a happy medium. This is the first time I've had a go at converting a machine and it's a fairly steep learning curve but it's at the point now where it's looking a little rewarding, I can't wait to have it up and running properly. I ordered a new shuttlexpress last night so it should be here in the next day or 2 and I'll get some wiring sorted tomorrow afternoon for the control box. I'm having thoughts about a tool turret as well now, it would make it a really useful machine but I think I'll get it up and running with the QCTP first so I can make some stuff.

Cheers,
Greg

[RotarySMP]

I guess having the hard stops at the end of physical travel is the standard way. My "new" lathe is a German Boley 4L for which the threading change wheels were not included. I am probably going to CNC it. It has a T-slot machined into the front face of the bed where movable ramps are mounted to throw out the half nuts at location which can easily be set. I am thinking about mounting end switches to it, so they are movable.

On the Chinese mini lathe I CNC'd, I did my Z-axis swarf guard nearly the same way you have, and it worked pretty well. Only different was a sloping curve to stop swarf building up on the roof.

I always used an AXA QCTP on my lathe and found it worked really well. Not as sexy as a turet, but much more flexible, and since I kept using it as an engine lathe, never making more than a few same parts, the manual tool change was no restriction.

I used TurboCNC until I got sick of the half-finished mix of Tool offsets and work offsets in one register. I then switched to LinuxCNC, will is a a steep learning curve, and lacks the cool canned cycles of TurboCNC and Mach.

Is that Leadshine a two axis or a three axis driver?

[proengines]

The more I think about it they may be a waste of time but if I didn't fit them it would crash just to spite me! I'm getting the cover folded up this week, there's enough room for a bit of a slope on the top of it, the slot is 13mm wide so I might do that.

I like the idea of LinuxCNC but Mach is pretty user friendly and at the moment I need all the friends I can get doing this. I do like the canned cycles and it's looking like there's not much need for a CAM package for what I want to do. I am trying out the new Autodesk Fusion 360 though which is looking pretty impressive for the price.

The driver is 3 axis, I just liked the idea of everything incorporated and could use the other drive for a tool turret if I go that way plus they're not overly expensive compared to buying the individual parts. It definitely tidies up a bit of wiring.

[RotarySMP]

Do You already use mach on The mill? I considered it as the canned cycles look great. Their is some linuxcnc development in that direction as well, but everything in the linux world is geeky and it seems you quickly land in a geek programing environment when you try to configure anything.

Mark