Welcome to LinuxCNC.org

Hello everyone!
I recently purchased a NOS Jet 13x40 lathe that had been in storage for a few years at a friends facility. I was a steal and it was what I was looking for: something I could convert to cnc without worrying about throwing all the gearing away.

For the servos, I'm going to use 3 phase ac units and for the controls I'm going to use linuxcnc w/mesa cards. I'm also going to use 1 micron linear scales for secondary feedback on both axis.

I am curios as to how everyone else does the x ballscrew. I'd like to do a preloaded setup, but I have a feeling I won't be able to fit a double ball nut preloaded screw like the Rockford units. I was thinking about a 400W servo for this axis, or should I do 750W?

For the Z axis, I was thinking about using a 750W servo direct to a 5mm lead 20mm rockford preloaded ballscrew. Does this sound ok? Should I go to 1.5kW maybe?

For the spindle, I was thinking about a 7.5kW Servo, belt drive to the spindle with a 3:1 reduction which would give me a rate 1000 spindle rpm which is the same the lathe has now. I'd like to use a servo proper, because in the future I'd like to add live tooling on a slant way to this setup.

I'm working on an 8 position automatic tool post that will also be for this lathe.

I appreciate any comments, opinions or advice on this.

Frank

How big is the lathe? For comparison the 100W stepper on the X of my 9x40 is very rarely a problem.
The 400W unit (especially a servo) is probably plenty, and might be easier to package.

Lathe cross-slides are often difficult to retrofit as there isn't a lot of room. You might find a rotating-nut solution packages more easily.

I would be tempted to hold off on the linear scales until you have the machine running, they may be unnecessary.
750W sounds like plenty on the Z, though I am not sure that direct drive is ideal, 15m/min is probably quicker than you need.

Again, 10HP on the spindle sounds huge for a 13" lathe. Conventional lathes of that size can snap tools with a 3hp induction motor.
(www.colchester.co.uk/product_detail.asp?ProductID=6)

Have you priced a 7.5kW servo and drive?

Hey Andy,

The main theme with this lathe is that I got it as a trade for 16 hours of my time on a weekend wrenching my friends Jeep fuel injection conversion, something I enjoy. I would have done it for free anyway, so to me its brand new lathe that I got for free from a friend. This way, if I spend a few grand getting putting all the bells and whistles for it, I'll have a new cnc lathe for the price of a new manual lathe. Plus, this is a Taiwanese lathe not the newer Chinese Jets. At least thats how I sold it to my wife

The lathe is 13" swing and 40" between centers. I turn mostly 316, 4340 and 4140, so I wanted something with good oomph as well as accuracy. I think from the sound of your setup, this may be plenty power for the axis'. On your cross-slide motor are you using any kind of reduction between the 100w stepper and the screw? What size motor and reduction are you using on your z screw?

As far as the cross-slide is concerned, I am going to take it apart this weekend to take some measurements and see what can be done. I've been playing around with the lathe and I'm thinking it really wouldn't be that difficult to make a new saddle for that little lathe with a wider cross-slide to make it more stable on boring operations. If I did that I could also add turcite B to the ways and cross-slide.

I do a fair amount of low tolerance one-off parts, so my thoughts with the scales was that they would help with the accuracy of the machine since the rest of the machine is new and tight. You have an excellent point though, I'll leave the scales for last to see if I really need them.

For cutting 15m/min is useless, but I like having a good rapid with good acceleration. On a worse case scenario I can just dial down the feed rates and the servo lasts longer since its not wizzing around at 3k rpm. At the very least it sure looks pretty when you see the axis moving at ludicrus-speed.

Here's my reasoning on the 7.5 kW servo: The lathe currently uses a 2 HP motor with the reduction box. If I am doing a single reduction 3:1 toothed belt drive to the spindle, I need to make up for the lost torque on the lower gear settings. After I ran the numbers I figured that a 7.5kW servo (or a 3ph motor with a vfd and encoder) would provide enough torque that I can run it from 50 spindle rpm to 1000 spindle rpm without any issues. The used servo I am looking at is rated for 3000 rpms continuous with 4500 rpms max. So, if I felt brave enough, I could crank it up to 1500 spindle rpms for finishing. The servo and the driver will cost me about $2500.00, but to me, its still within budget.

Frank

dangercraft wrote:The X (100W / 1.5Nm) is geared 1:1 with an 8mm x 2.5mm ballscrew. This is definitely a bit on the tiny side, but has so far given no trouble.
The Z (200W / 3 Nmw) is a 200W stepper geared down 3:1, and that is actually a little weaker than I would like as it can't really do toolpost drilling over about 10mm.
Which just goes to show that steppers are funny things and you need to look very carefully at the speed/torque curve.While it is a lot of work, this is what I would do if I was converting the same lathe again (and if it didn't need to do double-duty as a mill). I would make a CNC-specific saddle and a cross-slide with built-in bosses for a toolchanger. Iron castings are not all that expensive, I paid £25 each for the motor mounts for theY and Z on my mill. This also means that you can remove all the manual parts unmolested, and possibly even recover costs. You can at the very least make parts without having to re-assemble the machine several times like I did.It looks pretty until you crash into the tailstock because you can't get to the e-stop in time . I am just thinking that more push for drilling might be useful, and for that you would want to be gearing down.Any idea what the spindle bearings are rated at? Is it worth keeping a reduction gear for large-diameter occasional work, and having 3000rpm for hard-turning?
www.mmsonline.com/articles/taking-the-fear-out-of-hard-turning suggest that you would want 2500rpm for hard-turning 1/2" or 12mm material. I have used it with good effect even on my made-of-cheese Indian knock-off of a cheap Chinese lathe. picasaweb.google.com/1081645046564043805...#5611905035055922322

Andy,

Thinking it over again, I think you are right, 3:1 with a 750w servo may be slower, but I'll certainly gain in acceleration with is were most of the time savings in machining are realized anyway.

I have a few chunks laying around of 6" 4340 plate. That may be my first job for the VMC thats coming in. I also like the idea of making it from scratch, because, as you said, I can work the tool changer right into the design.

I really like the idea of using CBN inserts, what I may end up doing is in the future adding a twin clutch transmission with to belts for a high and a low range. As far as the spindle bearing is concerned, I am not sure, but 3000 RPM for a spindle bearing isn't much, on a worst case scenario I would just swap the original for a spindle bearing proper.

Tomorrow I get the VMC delivered, so either way I have to get that retrofitted first so I can do all my other stuff.

Frank

I retrofitted my 14x40 lathe last winter. I used kelling 850oz servo motors which are plenty powerfull. Both X and Z are ebay ground ballscrews with quality support blocks. IAs I recall, my x screw is 12mm diameter. The X screw fit in the stock castings with only minor rework. I used 3;1 belt drives for both axis. I designed for 200ipm but have it set at 100ipm max which is plenty fast. For the spindle, I replaced the stock motor with a 3 phase one and added a vfd, that worked out very well.

What I did do that I am not happy with is going with Mach3 for the lathe. I am finishing up another lathe project right now under linuxcnc and then the 14x40 is getting re-retrofitted to linuxcnc also.