Turning a 3' screw on a 9x12 lathe

That sounds like a pain, and for reasons you haven't considered as well as ones you have. Long thing shafts do not turn well - you would need to set up a follower rest on the carriage to support the work, and it's possibly going to hit burs from the partially turned thread, etc. Not fun.

Unless there's extreme personal need to do the whole thing yourself just to prove you can, getting the project working is best achieved by buying the threaded rod you need, or modifying the specification (change the nut) to one that you can buy. The accuracy and smoothness will no doubt be much, much, much better. Note that you can partially bore out the old nut and use a cast in place low friction epoxy (moglice) to make one from the new screw.

I have access to a Clausing 17x80, but it does not have a follow rest. As you say, turning a thin shaft without one is impossible.

Maybe I could get a description from someone on how one would mount on the carriage? I can build one easily enough, I just don't see where it mounts.

This is for a home built plasma cutting table. That's why I need the length.

Andy

Probably bolts to the side of the saddle below the cross slide, I'd put it on the headstock side though I suppose it could mount on the other and reach around somehow. You might have to take the saddle off and drill and tap mounting holes.

Do like everyone else who builds that project and buy rolled acme screws and nuts, they are cheap. You can even get motion-transfer grade ballscrews for fairly cheap.

You still get to play with the lathe as you machine the ends.

Oh, they not only whip when being machined, they can whip in use when rapiding around to reposition too. So I'd buy the largest diameters screws my budget and avaialable space could support. 5/8" minimum,

3/4", 7/8" even 1" for the lower axis.

I'd have to agree with the other responder. Acme threaded rod is too cheap to go through this agony.

Just buy the acme threaded rod. I was reading my

catalog this morning, 1/2" ACME by 3' was $4.59 on sale. You can't beat that doing it yourself.

1/2" is probably too small. Eulers theorm says that if the length is 89 times the diameter it will buckle. Safe design would trim that down to the 30 to 50 times range. On a threaded rod you use the minor diameter (bottom of the threads), 1/2" acme has a .375" minor diameter so 36" shaft will be at a ratio of 96:1 WAY to spindly.

TheAndroid wrote:

see

Uncle George

What you need is a Ball Screw . I think I have one that long in my junk pile.

Bill K7NOM

It would be awesome if you had two!!!!! I would be very interested!!!!

BTW, thanks everyone for the input. I looked at threaded rod, but multiple start rod from McMaster were 49+ dollars for 6 feet. I was hoping to trade some time for cost on these screws. Additionally, the steppers I have are rated at 240 oz/in in bipolar mode. I wanted to minimize the torque required to turn the screw. I think a 1"" screw at 3 feet would simply weigh too much for these motors to turn and stop at reasonable speeds.

As far as follow rests go, the links provided have given me much to think about. I will probably move ahead on building one as you never know when the need will come up!!!!!

Andy

Do you have a reference for Eulers therom? This sounds like something I may need to research a little more before I start building.

The one for my Clausing (12x24") mounts to the flat tops of the extensions to the left of the carriage. There are threaded holes with studs in them. The follower rest drops over those and is secured by nuts run down onto the studs.

Good Luck, DoN.

Hmmm, I was debating if I should use a 2 inch dia ground ballscrew (8 feet long with ballnut, brand new for $125...gotta love Boeing Surplus!) on *my* gantry table design....it seemed rather large, but I guess it woudl cut down on the whip...and I have it...kinda a shame to use it on a machine that will probably only cut wood...

Why do you want multiple start threads? I think the usual use is to have a faster lead than the apparent groove pitch so that high powered systems can rapid quickly.

Which is probably not aided by a multiple start thread?

Though I do think there is something about with ordinary screw threads, a finer pitch does not actually get you much mechanical advantage as friction dominates. A cheap ballscrew on the other hand...

You may want to look at toothed timing belts for connecting the motors to the screws. This can let you change the ratios experimentally, but it also introduces a nice solution to the problems of misalignment and vibration damping between the motor and screw. Without a belt in there, you'd need a flex coupling.

According to Rick :

One possible approach to minimizing whip is the one which Bridgeport used for the X-axis on their CNC machines -- starting as early as the BOSS-3.

They simply rigidly mounted the leadscrew to the table at one end, and instead rotated the ball nut in paired bearings. Of course, it made manual operation pretty much impractical without the computer aiding in a sort of "fly-by-wire" technique, as the nut was down under the table mounted to the saddle along with the motor.

But a skinny enough and long enough leadscrew will still develop resonances at the right (or wrong) frequencies, so fatter is better. You just may have less mass to accelerate rotating the nut instead of a fat long leadscrew.

Enjoy, DoN.

Note that *some* types of garage door openers have triple-start Acme threads. And the ends have 'D' shapes (IIRC) to allow them to be coupled together in sections to adapt to the needed travel of the garage door.

They run in aluminum extrusions, and the "nut" is a plastic one long enough to bridge over the coupling gaps.

The maker said that shorter sections were not available (I needed one to clear the enclosed I-beam supporting the next floor), and when I suggested that I would just have to modify the existing ones, they said that it was not possible. :-)

Of course, it was possible, if you have a lathe with a big enough spindle hole to accept the OD of the rod -- and even my Compact-5 was big enough for that. :-)

Good Luck, DoN.

Sorry, I didn't explain fully. I get going sometimes and miss details......

Multiple start threads allow me to get the faster lead without extensive thread depth. I wanted to cut the 2 start 10 TPI because the

10 TPI thread depth would leave me much more root diameter of the screw (vs the 5 TPI depth) thus hopefully reducing flex. My real target is 5 TPI as my steppers are 200 steps per rev. That gives me a .001 resolution. I can drive my steppers at about 150 pulses a second without missing steps. This gives me around 9" a minute top speed movement. This is a bit fast for plasma cutting heavy material , but keeps me well within the capabilities of the steppers without introducing potential backlash other than from the leadscrews. I plan on making anti-backlash nuts for those.

Even so, I want to reduce the rotational mass the steppers are turning. The suggestion by Don is very interesting! I plan on having a laptop control for the table anyway so I don't care if I can manually place the cutting head. I need to see if I can find how they drove the nuts.

Andy

Thanks for the help everyone! I had posted a reply to the question about why I needed multiple starts, but apparently it fell into the bit bucket.

I'd like multiple starts because it allows me to have 5 TPI lead without the thread depth of 5 TPI. This leaves me with more root diameter of the screw which will presumably give me more strength. My stepper motors use 200 steps per revolution. With a 5 lead, I get .001 per step resolution. That's really the aim.

Andy

My replies keep getting eaten!

Thanks for the help, everyone!

I want multiple starts in order to increase the root diameter remaining of the rod after threading. 2 starts use half the thread depth.

Andy

My replies keep getting eaten!

Thanks for the help, everyone!

I want multiple starts in order to increase the root diameter remaining of the rod after threading. 2 starts use half the thread depth.

Andy