Newbie Question: Lathe Leadscrew Handwheel

[J Tiers]

thats what mine was. As is my horizontal mill, its just not an issue. For all the naysayers - ever had a lathe with one? Until then.....

....

Yes I did.

It was a big advance to move up to a regular rack feed, the more so as that machine ONLY had the hand wheel or halfnuts as options. With both the rack and the handwheel, might be OK. I have thought about putting on a fine feed attachment. I would NOT go a handwheel.

The point was not that a handwheel is BAD, just that it is not the only, nor even the best, way to go. If it was so good, it would be a standard feature. But it is not, for the vast majority of lathes ever made.

It is less accurate than a DRO, less accurate than a micrometer stop. Has a LOT of places for slop and backlash that need to be taken up, and which may kill precision unless all operations are done at one closing of the half nuts. And the dial calibrations are a problem for nearly any imperial leadscrew other than 8 tpi.

Could be pretty nice as a fine feed, though. THAT feature was good.

[lakeside53]

Compact 10 - has the same hand wheel also.

[Mcgyver]

Emco 8.6 and V10 have an 8 tpi lead screw - the hand wheel has 125 divisions in thou... You get used to it very quickly.

thats what mine was. As is my horizontal mill, its just not an issue. For all the naysayers - ever had a lathe with one? Until then.....

It is accurate to a thou, you're not doing it "all the time". Its a feature that is handy occasionally when say turning to multiple shoulders and I also recall using it to turn radii (x/y method). I also regularly used a micrometer on this lathe (now that is needed accessory imo) so it was not intended to replace that.

[Paul Alciatore]

My SB has an 8 TPI lead screw. That means 0.125" per revolution. I have thought about using a stepper motor on it with some timing pulleys and an appropriate ratio to work with that.

For instance if I wanted to be able to move it by tenths, that would need 1250 divisions per revolution. Now, most steppers have 200 steps per revolution so that would mean a ratio of 6.25 (1250/200) for the pulleys. That's kind of high but it could be done with 8 teeth on the stepper and 50 teeth on the screw. Then each step would be 0.0001". Working in a control knob with 100 pulses per rotation would give you 0.0100" feed per revolution. That means 100 turns for a one inch movement. Gonna be a lot of cranking. Perhaps the hand wheel would need a 10X and even a 100X speed feature(s).

I could add some burst controls. Push buttons that would give you some whole number amounts of feed like 0.2500", 0.500", 0.7500", 1", 2", etc. These would need a variable frequency capability to control the feed rate if you were cutting. Could even add a thumbwheel switch to allow settings from 0.0001" up to the length of the bed. That would require a six gang switch. Set it, push GO and it does the cut.

All this would need some work and experimentation. These are just wild thoughts at this point. I am sure that some of these numbers would change as the idea develops. Perhaps I could settle for a minimum motion of 0.0002". It would take some real thinking to determine just how something like this could be best set up for use.

And it could even have some metric controls with the appropriate logic between the controls and the stepper. Now there's a thought, an English/metric lathe with a simple switch to choose which one you want.

PS: I don't want to here about half steps or microstepping unless you want to explain just how large the stepper motor would need to be to maintain decent accuracy with that. Do check out the derating curves before even thinking about suggesting that. Personally, I do not like microstepping with a stepper.

[lakeside53]

Emco 8.6 and V10 have an 8 tpi lead screw - the hand wheel has 125 divisions in thou... You get used to it very quickly.

[benjaminh]

Thanks for all the input guys! The points about problems with dial division due to leadscrew tpi is a good one that I never thought of. I think I'll forget about this idea for right now, and see how using DI's and stops works. I can always add the handwheel later if I feel I really need it.

[Paul Alciatore]

The use of an uncalibrated hand wheel for moving the carriage is not completely a bad thing. I have a small lathe, a Unimat, and it only has a hand wheel to move the carriage. The carriage movement is only about 150 mm or 6 inches, but at 1 mm per revolution, it takes a bit of time to move it from one end to the other. The hand wheel on my SB-9 if a lot faster for rapid transverse than any hand wheel would be.

Another common solution to this situation is to mount a DI on the ways where it can measure the carriage movement. I have a 2" one for my SB-9 and I find it very helpful. It does not have the range of a DRO, but it does cover at least 90% of the jobs that I do. For longer distances, I would do a rough cut and measure the part. Then zero the DI at that point and remove as much more as the measurement calls for. This works well.

You can see this 2" DI here.



I have made a new mount for it since this photo was taken. The one in the photo came with the lathe and was rather shabby and shaky. The one I made is a lot more solid.

I like to save my lead screw for threading. But sometimes I use it for power feed. I need to upgrade to the SB-9B model.

[darryl]

Constantly driving the carriage with the lead screw would drive me nuts. As far as the hand wheel that drives through the rack, I removed my hand wheel and replaced it with a section of aluminum tube about 2 inches diameter and a hand length long. I find it much more convenient, and it's easier (for me at least) to make small adjustments to carriage position- it's all about feel. If I need a fast traverse I just push the carriage while the 'hand tube' just spins.

What I've done that is more in line with a handle on the lead screw is to add a section of 10 tpi threaded rod to the right of the carriage. The rod has its own adjustment wheel which is calibrated, and it's mounted in a fixture of its own which can be clamped to the ways at any point. If I leave it loose I can move the carriage using the rack, and when I want fine control I clamp the fixture and use that wheel. Of course, it takes a full turn of that wheel to move the carriage 100 thou, so if you rely on that for anything other than fine control it becomes very tedious.

[J Tiers]

While that can be a nice feature, it can also be less useful than it seems.

The leadscrew drives the carriage through the halfnuts. The halfnuts have slop between them and the screw. The halfnuts also have slop between them and the frame that they ride in, they can move sideways, or tilt a bit. The leadscrw itself is in bearings, and has some endplay.

But I noticed Mcgyver's handwheel had a dial for calibrated feed. Maybe that is fine with a good, tight machine. But with multiple sources of looseness and slop, it seems pretty optimistic to count on the screw as a source of accuracy down to the thousandth of an inch. For threading, with the halfnuts closed for one continuous "pass" over the part, it is reasonably well compensated-out or "taken up". For any location of multiple features, with the halfnuts opened and closed, carriage moved back and forth etc, there are likely to be a few thou of error here and there. Maybe good enough, but perhaps not.

And, if you have a leadscrew with an unusual pitch, the dial nay not turn out right. A 6TPI leadscrew is not that odd, but does not make for a nice set of divisions on the dial, at 0.166666.... inches per rev. An 8tpi screw is fine (0.125 inch per turn, but 16 TPI, which is fairly common produces 62.5 thou per rev, another issue for any measurement that involves more than one turn.

It is a rare machine with a 10 tpi leadscrew that would give 100 thou per turn. (there are metric equivalents to the issue as well)

Of course, if all you want is a nice slow hand feed, neither issue is a problem for you.

A DRO, or a micrometer stop, a "Trav-A-Dial", or the like will give a precise indication of ACTUAL carriage travel, totally independent of the theoretical travel per turn of the screw, or rack feed handwheel.

One other issue with the feed via the screw, is that habitual use of the screw for feeding will end up wearing the screw heavily in the most common area of use, leading to inaccuracy, and a "knuckle" in the travel where the worn area ends.

[lakeside53]

Emco 5, 8 and 10 inch lathes have a hand wheel on the lead screw. It's useful for many things needing fine adjust, but particularly when you have their mill installed.

I'd love to put one on my current lathe but it would be about 6 feet to the right of the chuck

[J Harp]

Several years ago there was an article in either HSM, or MW magazine about adding a handwheel to a lathe. I recall that the author was coming up with half a division when graduating the wheel.

[Mcgyver]

one of my earliest projects.....from one of Lautards books. First photo is from '97, second in 2010 when I sold it. I was an great lathe....but more just kept arriving

btw, myfords and maximats (the 7 anyway) had these so they were not uncommon. it is a nice feature. The alternative as, carriage stop and compound or better still micrometer stop are good but this feature let you turn to several shoulders.

[brian Rupnow]

By all means, add a handwheel to the end of the leadscrew. My first lathe, a B2227L from Busy Bee had that handwheel, and I loved it. When I sold that lathe and moved up to a larger one, the larger lathe didn't have that handwheel, and it was a miserable thing to move and position the carriage with that big handwheel on the front of the apron. I ended up spending over a thousand dollars to buy a DRO system, mainly so I could move the carriage using the power feed, and have some idea of how far the carriage was travelling. I would add that handwheel to the leadscrew in a minute!!---Brian Rupnow

[J Tiers]

...This particular lathe has no handwheel on the leadscrew, but rather only a wheel on the saddle that engages the rack. This obviously has no graduations, and is difficult to move precisely. I've noticed a couple other smaller lathes that are set up the same way.....

Just about every single lathe in the world is made the way your Granville is made. Some even gear that for threading, so they can do native metric and inch threads, one via the leadscrew, the other via the rack.

The use of a leadscrew handwheel is very rare, and normally is on tiny machines that may not even have a regular handwheel. The old cheap Sears "109" is like that, it has ONLY the handwheel, no rack, so you have to move it manually for any sort of rapid positioning.

That's why many machines may have a Digital Readout system (DRO) or a "Trav-A-Dial" that reads out distance on a dial.

[danlb]

Yes, in general the topslide is used for precision positioning on the horizontal axis. A common practice is to set a carriage stop, then use the topslide to dial it in precisely. After that you can use the handwheel to shuttle the carriage back and forth quickly while roughing it out or slowly for a finishing cut. The carriage stop will limit the travel.

You get used to using the carriage handwheel in order to move the carriage slowly. There is a lot of control available with a 6 inch wheel that turns a small gear which engages a rack.