Lapping thread?

[Rotate]

When you machine a screw and it's to be used for providing smooth axial motion, does it make any sense to use fine lapping compound with the nut so that the mating surfaces are polished? Is this ever done or am I dreaming in technicolours again?

Albert

[Oso]

Assuming you mean to lap the two together, not such a great idea.

One problem is that the clearance will always be about equal to 2x the grit size.
Has to be or the grit won't fit in.

Then too, if soft, there will be grit embedded which will continue to lap the harder part forevermore. Of course then the size will be closer, since the grit can bed in.................always a good side to things, eh?

[This message has been edited by Oso (edited 09-16-2002).]

[Thrud]

Albert
Precision lead screws are ground and lapped - and then fitted to a precision ground recirculating ball nut for essentially zero lash.

[SGW]

There is some information about lapping leadscrews in "Foundations of Mechanical Accuracy" by Wayne R. Moore of Moore Special Tool Co. What Thrud refers to. But they do the lapping in a controlled way to achieve particular dimensions.

I think Oso's right -- simply lapping screw + nut together isn't likely to work very well, if you're after a high precision fit.

What I've done to polish a thread is get a soft pine stick, dip the end in some fine lapping compound, and press that against the revolving thread. The wood quickly conforms to the thread form (assuming a relatively fine thread pitch) and holds the lapping compound quite well. I wouldn't necessarily call this technique "high precision" either, though.

[hms50]

Hi Rotate,

I've lapped damaged threads this way with good results. I submitted an attempt at an article to Neil about this which should be out shortly. It's a trick much used to repair antique pocket watch cases when somone has forced a case-back off without unscrewing it. It will improve the fit quickly and make it feel super smooth. In the case of a feedscrew, I don't think it would be a good idea to use the nut as a lap. Iv'e made up an adjustable lap of a soft metal, concentrated on lapping the tight spots on the screw and used a new adjustable nut when completed. This way, the grit is thrown out with the lap and can't continue to lap the screw. I don't know how to remove grit from a lap completely enough so it won't continue to cut. A Briggs & Stratton rep told me this is why it's hard to find anyone willing to "bore" al cylinders. At the factory they have cleaning machines of some kind that does the trick but they are much $.

As to how accurate the resulting screw will be,the only testing I did consisted of mounting a dial indicator and comparing readings to the micrometer dial on the feedscrew. The results were fine for a lathe crossfeed, but for real high class work they might not be.

Good Luck
hms50

[Rotate]

I was thinking of lapping the screw not to improve the fit or to reduce the backlash. If anything, I think it would worsen it. I was after a very "smooth" motion transfer so that the friction of the rotating motion would be consistent, devoid of minor burrs.

I find that the screws that I cut always shows machined marks on the thread (under x10 loupe). So, I guess the follow-up questions is, how does a HSM grind the screw and the nut?

Albert

[Bob Quale]

I had a brass half nut that I lapped against a lead screw. It worked well and still does, however I am wondering how long it may last. I never thought about the lapping compound getting into the brass and continuing to cut. I can't see any significant wear on it yet, but in the home shop how much use does it really get.

BOB

[Uncle Dunc]

My first impression is that the five or eight or whatever turns of engagement would work to average out any roughness in the thread or the nut. To get any roughness in the screw motion, there would have to be some kind of regular relationship between the distribution of the tool marks on the screw and the nut. For instance, if you had the same number of chatter marks per revolution on both, you could feel them cogging past each other as you turned the screw.

What's the application? On machine tools, at least all the machine tools I've worked with, when the gibs are adjusted to minimize side shake, the friction of the ways pretty much swamps any delicate feel on the leadscrew.

[Rotate]

The application is to tilt a mirror for an optical experiment. The screw is 1/4" fine thread and is driven by a small DC motor with a 100:1 step down gear. The mirror is spring loaded so the screw and the nut are always touching on the same side so backlash is not a problem. The experimenter uses two switches to control the tilt. I've played with the unit and I find that there are spots where the angle of the mirror skips. Needless to say, this skip is extremely tiny.

How about running the screw on a buffing wheel with rouge compound?

Albert


[This message has been edited by Rotate (edited 09-17-2002).]

[Uncle Dunc]

How long is the engagement between the screw and the nut? If it is only 2 or 3 turns, the first fix I would try is a longer nut, followed by a stiffer spring, followed by polishing the screw. I don't really think the screw is the problem, but it might be, and it's an easy fix.

If that doesn't work, I'd start trying to characterize the skips. If you see it skip and back up for another try, does it always happen in the same place? How many revolutions is the total travel of the screw? Can you tell if you get skips at the same point in each rotation? If you do, is it over the whole travel of the screw or only part of it? If it's only part of the screw, then you've isolated one area that needs closer scrutiny. If you do see a repeated pattern of skips, can you turn the screw a quarter of a turn relative to gear box output shaft to see if the pattern follows the angular position of the screw or the angular position of the gear shaft? If you see a repeatable pattern of skips in one rotation, do you see the same pattern coming and going? If you do, is there an angular offset between the two directions? If there is, that again would suggest that the problem is with the gear box.

Of course if you've already polished the screw and find out the problem follows the screw instead of the gear box, I don't know what to suggest.