Shop Fox M1098

Have you looked at buying a dead CNC lathe? It does take searching but you'll be way closer to your goal. I'm partial to Hardinge because so dang many were made twenty years ago.

That said, looks like a decent enough import lathe. I'd have a design concept in mind for your ball screws and stepper/servos before you buy.

Karl

Karl Townsend fired this volley in news: snipped-for-privacy@4ax.com:

I have looked, Karl, but two things stop me. Most of the used CNC machines I'd have access to (say, in a 200 mile circle of me) are darned-well USED, and not in stellar condition.

Also, I cannot do the tax re-capture single-year amortization of a used machine. Buying new gets me an ROI of about 16 months.

The guys at Mach Motion have already done a nice conversion of exactly the same model lathe of a different badge. I'm not overly impressed with the Mach3 control, which apparently has trouble doing threading unless you buy the (additional) MachMotion $2550 stand-alone motion controller that goes between Mach3 and the drivers.

I'm pretty certain that Iggy has proven EMC will do it without all that additional computing power and unsupportable embedded proprietary code.

However, since they've converted a lathe that is mechanically identical to the Shop Fox, I assume it's not to ugly mechanically -- and heck! I'll have the lathe working in manual mode to do the work .

LLoyd

How about starting with a Lancer lathe? i see a fella here offering a project lathe cheap. And you wouldn't have to take it apart first I bet he'd find you some ball screws and servo motors to sweeten the deal.

Seriously, that would make one hell of a lathe.

Karl

I've got a better accountant than you. No problem to section 179 used equipment and expense it.

Karl

Lloyd, my opinion is worth $0.01, but I feel that I have to state it.

CNC retrofit of a machine that is built as a CNC machine, is a COMPLETELY different type of project than a CNC retrofit of a manual machine.

If the difficulty of the former is 6 on the scale of 1-10, then the latter is 9-10.

You have to be essentially a CNC machine designer, make a lot of things fit that do not want to fit, make and mount a lot of brackets, doodads, switches. Also you need to buy servo motors.

It all seems simple conceptually, until a) you start doing it and b) make some mistakes that you realize late in the game. For sure it is NOT cheap, if you price your time at $5 per hour and up.

I agree that a CNC lathe that is not totally clapped out, is harder to find that a mill or a VMC.

I may be overstating my case or just not experienced enough to breeze through that sort of thing. But somehow, I feel that retrofitting manuals machines is not nearly as cost effective as taking a CNC machine with a bad control.

I've followed the Mach3 forums for years and I've not heard of anyone reporting problems with threading. Sounds more like a sales pitch to sell the extra hardware.

Mach3 is based on EMC. Mach3 is also well supported.

Always good to be able to follow an existing design vs. rolling your own design.

Ignoramus18335 fired this volley in news:68edneNExM3Yz0zQnZ2dnUVZ snipped-for-privacy@giganews.com:

Ig, I take your advice seriously, but I have the advantage that they have already converted one of these. Only two small issues must be resolved on my end:

I must make a ball nut bracket for each of the Z and Y feeds, and I must add thrust bearings to the ends of the screw brackets to take out end play. They have motor mounts and couplers designed to purpose, and the adaptor plates and couplers only add about $50 to the total bill -- saving me much layout and machining time.

I have pretty much exhausted the near-regional used CNC lathe market looking for anything that doesn't have so much wear as to require a full screw and motor upgrade (and way scraping), and I think I'd rather do the conversion. I intend not to kill the manual functions until everything is ready (although I'll probably do several disassemblies and re- assemblies of everything before I'm done).

Things like mounting limit switches are just not that much of a challenge.

As for mistakes? Yeah... but they're all "recoverable", so long as I don't invest in motors or ball screw assemblies that are sub-standard. To that end, I'll be using some proven motors, and precision-ground screw/nut combinations. The precision of rolled screws just leaves me cold -- 0.003" per foot is crazy sloppy.

LLoyd

Gunner Asch fired this volley in news: snipped-for-privacy@4ax.com:

I need a pretty good-sized work envelope. That's one reason I'm going with a 13x40 manual.

What models do you have, how much, what condition of mechanics, and how much for shipping one to Florida?

LLoyd

Keep us posted Lloyd, I do not want to discourage you in any way.

Ignoramus18335 fired this volley in news:w8mdnTXWPd4DPUzQnZ2dnUVZ snipped-for-privacy@giganews.com:

In any case, this will be pretty rapid-fire, except, perhaps, for the budget. The motors and drivers will cost me $1200 for new, proven AC- Brushless Servos and 2500 line encoders. That's the biggest rub. The screws are pricey anywhere you get them, when you go to the 0.0005"/ft. versions.

I'll be working hard to preserve all of the manual features except for threading (I have a 6x18 that can do for the time being), until that very last moment when I "switch over".

LLoyd

I've only seen two used lathes in my area in two years, and you wouldn't have wanted either of them. At least one was scrapped because it needed so much work.

Gunner Asch fired this volley in news: snipped-for-privacy@4ax.com:

And that, plus heavy-duty shipping. That's why I'm keeping my look to a

I sold my F.E. Reed and Cincinnatti #2 to a guy three hours away, and he took them grinning. But, then, the only diseases they had were pure age

-- ways scraped, bearings running close and smooth.

LLoyd

Old Okuma cnc lathes are very good candidates for retrofitting. Find one with an old OSP 2000 control and you will have a very good start. The biggest problem is that they used absolute encoders which were great on the original lathe setup. No homing was necessary.

John

The option is to use glass scales with a rolled screw. AS long as the ball screw has no backlash the glass scale will give accurate position with no backlash.

John

Don't forget that CNC controls include screw mapping / pitch error compensation. Measure and map the screw once and all the inaccuracy is gone for good, or at least until you wear out the screw. Commercial machines / controls use screw mapping as well.

"Pete C." fired this volley in news:4dd28edf$0 $10550$ snipped-for-privacy@unlimited.usenetmonster.com:

Yes, true enough, but the one thing I don't have is the gear to measure that over 30-some inches of carriage travel. On the one hand, the hardware would be cheaper, on the gripping hand (using precision screws), I wouldn't need the instruments (I can live with a thou. over three feet). Maybe I could get someone who does have the tools to come do the mapping...

LLoyd

Most people use a dial indicator and map the screw an inch at a time.

There are calibration service companies that will do the job with laser interferometers. There are home-made solutions, but they're very tedious and slow. And I forget what they are. d8-)

"Pete C." fired this volley in news:4dd2914a$0$10550 $ snipped-for-privacy@unlimited.usenetmonster.com:

Not hardly, Pete. I've watched that procedure before on big VMCs. You want not only "localized" errors, but cumulative errors to be mapped.

As Ed stated below, the Japanese techs were using laser devices. The could map a 6' travel to half a tenth, cumulative, easily. (I'm betting it was better than that)

LLoyd