Would you convert to CNC a 13x28 SouthBend lathe with around 0.010 wear, top speed 940, thread spindle?

[tkitta]

Would you convert to CNC a 13x28 SouthBend lathe with around 0.010 wear on the ways? It has a screw on spindle and top speed on plain bearings is 940 through people have run it 1200 rpm without any issues & even some advertise 1600 rpm (but probably with maybe a forced pressure oil on the plain bearings).

The lathe is worth about 2000 CAD (I am in Canada) and conversion to CNC would cost me another 2000 CAD at least. Plus a lot of work. So would the end result of spending around $4000+ be worth the effort?

Cons:
wear on ways + looks kind of ugly
slow spindle speeds
threaded spindle (my other machines are D1-3 and D1-6 so I may have to build a thread - D1-3 adapter)
Pros:
base machine is cheap


Thanks for your thoughts.

 

[sdelivery]

Probably not. Modern tooling and machining techniques require more sfm or spindle rpm, especially with small diameter work.

 

[graham-xrf]

With all lathe leadscrews, there is backlash. A small amount of backlash is even necessary for humans to know and trust the move they make has some certainty. For CNC, we are now talking servo feedback control, for which backlash is untenable. If a servo motor drive starts moving, even by microns, and the feedback is still not reporting movement, the drive energy racks up rapidly into an instability oscillation that can rattle your teeth!

Step #1 for any any conversion to CNC is to replace the leadscrews and nuts for a ballscrew drive which has zero, or near zero backlash.

Adding CNC to a ready-made lathe infrastructure gets you end up with a sort of CNC machine. You will have invested in the CNC hardware, a real amount of adaption design and learning, and the time and effort to install and calibrate. For many at HM, much of this would be very cool fun to do, and succeed at. The added value to you is that in getting to that point, you also end up with a deep understanding of CNC and your machine. This state is different to one who can operate a CNC machine by training.

I am not sure what value one would put on undertaking and completing a conversion + conversion hardware costs. The raw cost of a purpose-built CNC machine is likely to exceed $4000, even a used one, and you would not know exactly how it works, and would still require the learning curve to use it.

CNC adherents in this community do all kinds of conversions and also "from-scratch" CNC builds. Some will purchase linear bearing precision slides, and ballscrew drives, and make their own stable beds from steel reinforced epoxy-granite in home-constructed plywood molds. The machines are often individual, unique things of beauty. A regular machine, without CNC, is designed around the fundamental philosophy that is part of a scheme that includes an advantageous, built-in, twelve billion cell human brain (+ other soft parts liable to damage). True CNC, taken to the point it's strengths are properly exploited, involve higher powers, speeds, and motions a human cannot do, + they never stop to bring out the micrometer and make a measurement.

It's interesting, and looks like fun, and clearly we have a load of experts here, but honestly, somewhat out of my league right now. I do aspire to motorize the cross-feed on one of my lathes, and I already know I can use a servo drive main carriage leadscrew instead of using a set of change gears. Not true CNC, but easy enough to do.

 

[mattthemuppet2]

I'd say no. Easier and cheaper to buy something closer to what you want. Not alot of smaller machinery up there I'd guess, but you could probably do an epoxy/ granite base and linear rail build for similar or less money (and effort!).

 

[MrWhoopee]

No. That machine may be capable of producing good parts with a skilled hand monitoring and adjusting for its shortcomings. It will never produce good parts as a CNC, which assumes that the machine does exactly what the controller tells it.

 

[spumco]

No.

You can upgrade everything, re-scrape the ways, add world-class ball-screws and a million dollar controller... but the spindle is what will cripple that machine.

It has a tiny through-hole and a threaded nose. The threaded nose is unsuitable for CNC - you can't run your tools upside-down for chip evacuation, can't run in reverse (safely). You would wind up with a CNC machine you have to babysit the entire time.

 

[macardoso]

If you really want to go the CNC route, look for a used CNC machine in good condition that is within a weight you can handle and put some modern controls and motors on it.

The machine is already designed for ballscrews and motors, and will likely have a chip and coolant enclosure which is a big plus.

I found a 13x40 lathe about a year ago that was full CNC with an enclosure. Might have even worked fine. $1500. Even if the controls were dead that price is an excellent place to start on a build. Wish I had a ground level shop to have bought that.