question and first post

[motorworks]

Hi
Been looking around for a while, great site

Quick question
Building a self feeding line-boring bar and I was wondering if
something like this differential screw would work as a feed?
see link:

[ame]http://www.youtube.com/watch?v=iPkoGug-oC0&feature=related[/ame]

I was/ am planning on using an inline speed reducer and a screw with a fine pitch thread (3/4-27) on the end but I came across this
and it looks more compact.
Anyone ever built one?
Any thoughts

 

[cfellows]

Can someone help this befuddled brain understand how this works?

Chuck

 

[raggle]

Can someone help this befuddled brain understand how this works?

Chuck

No!

But here are a few observations. The male thread being considerably smaller than the threaded hole in the bearing has a rolling action within the bearing. It is constrained from losing engagement by the next bearing being "in opposition" then the third taking the same radial position as the first.

Quite how the motion is transferred from one bearing to the next is puzzling me at the moment, except for the feeling that "something's gotta give" somewhere. A single bearing would just idle ???

The guy in the video says it converts 32 tpi into an effective 100 tpi (could be he meant 96 tpi, but I will gladly stand corrected) and if I'm correct it is a function of the number of bearings. Or could it be the relationship between the shaft diameter and that of the bearing inner? Then I wonder if adding more bearings increases the effect.

Gut feelings are it "resolves" the motion and would be devoid of backlash.

Quick question
Building a self feeding line-boring bar and I was wondering if
something like this differential screw would work as a feed?

I feel sure it would! Just wondering how you tap a hole in a bearing but maybe they're available? A question I've never thought to ask.

Looks like me and Kludge could get a 0.010" per rev feed on our Taigs driving a leadscrew 1:1.

Think we need Marv on this one :bow:

Ray

 

[Stan]

I pondered this when it first appeared on the BB. The best I came up with was to look at as speed increaser (instead of decreaser as shown in the video) by holding the bearing and turning the shaft. I can visualize the bearings acting like a multi start thread.

If it does act as a multi start thread, then it would slow down the motion as he shows in the video by holding the shaft and turning the bearing.

Does this make any sense?

 

[GailInNM]

I will take a stab at explaining it without drawings by taking the principles one step at a time.

First visualize a rotating shaft, say 1/4 inch diameter. Next place a ball bearing of 1/2 inch bore on the shaft. Press the ball bearing down on one side of the shaft sor the pressure causes the shaft to rotate the inner race of the ball bearing. The inner race will rotate in the same direction as the shaft driven by friction of the staft to the inner race. The inner race will rotate at a ratio of the shaft diameter to the race ID. Using the numbers above the race will rotate at half the speed of the shaft. The closer the two diameters are, the closer the rotational speed will be.

Now put threads on the shaft and the inside of the bearing. The threads have the same pitch, but different diameters. If the threads are perfectly formed, they will roll against each other at the ratio of the pitch line diameters of the threads. If the nut were held still and the shaft make one revolution, the nut would move a distance of the pitch of the threads as the threads slide on each other. If the nut is allowed to rotate the half turn created by the friction between the nut and the shaft, it will move a distance of 1/2 the pitch of the threads, so the relative distance between the nut and the shaft will be 1 pitch ninus 1/2 pitch, for a net movement of 1/2 pitch.

As the ratio of the nut part and the shaft part decreases, the nut moves closer to the rotational speed of the shaft. This increases linear ratio of the thread pitch to the linear travel.

Math:
Let d = the diameter of the shaft at the pitch line of the shaft
Let D = the diameter of the nut at the pitch line of the nut
Let P = the pitch of the thread

Then tlinear ratio = (1- d/D)
and the linear movement = (1-d/D) * P

As a practical matter, the system is not good for a lead screw if you fit a dial to the shaft as the ratio is affected by any slippage in the friction portion of the drive, and wear in the shaft or nut will change the ratio. For light duty positioning where the movement is measured by an external means it will work well.

Gail in NM,USA

 

[rickharris]

Two questions strike me:

1. How did he tread a hardened inner race?

2. Why doesn't spinning the outer shell simply cause the shell to stay on one place and rotate on the bearings. Bearing drag? If you load this wont the bearings just spin?

 

[GailInNM]

Rick,
1. A threaded sleeve can be inserted into the inner race on the bearing. Press fit or Loctite.

2. The center line of the bearing is not the same as the center line of the shaft, therefore the bearing orbits the shaft as the outer sleeve is rotated. Since there is pressure between the shaft and the ID threaded portion, the ID portion rolls around the shaft giving the relative motion as the threads mesh.

Gail in NM,USA

 

[Kludge]

Looks like me and Kludge could get a 0.010" per rev feed on our Taigs driving a leadscrew 1:1.

Huh? What? Did someone call?

Seriously (or as seriously as I usually get), I've been watching this & scribbling notes as quickly as I can ... though I keep having to sharpen my fingers & I'm rapidly running out of forehead. The basic idea's beginning to sink in but I'll have to sit and stare at it for a while before I have a useable degree of comprehension.

On the other hand, if it works as I think it works, then creating it using low peasant cunning might not be so difficult.

Best regards,

Kludge

 

[Dhow Nunda wallah]

Likewise, scratching head.
I've used differential threads as a static adjuster but you've lost me with the dynamics
???

 

[raggle]

In the spirit of acknowledgment I'd like to say thank you Gail in NM for the explanation :bow:.

I seem to recall that Leonardo described a system of resolving two similar, but crudely made, threads into one more accurate thread. Further iterations using these newly cut ones improved them further.

I'm sure its main application is in fine adjustment and I'd hesitate to use it for screwcutting. It may well suit as a fine feed on a light lathe like the Taig which doesn't already come with a leadscrew. If your length of 10-32 studding wore out I'd imagine it ain't expensive. Could be driven by a belt or a multi-speed dremeloid. Reversing out for the next cut, however, could take all day

How any of this helps Motorworks in his quest I have no idea.

Ray

 

[motorworks]

Thanks for the replies so far:

"As a practical matter, the system is not good for a lead screw if you fit a dial to the shaft as the ratio is affected by any slippage in the friction portion of the drive, and wear in the shaft or nut will change the ratio. For light duty positioning where the movement is measured by an external means it will work well."

I plan on using it to "feed" a line boring bar, it does not need to be an accurate movement, but it does need to be able to 'push through' the cut (0.025" deep)
Hoping to get a 0.002"/Rev feed or less.
The bar will be 3/4" (19mm) and the bearing bosses I am boring are approx 75mm thick (3")
So I was wondering if the"slippage" would be a problem
take care
eddie

 

[GailInNM]

Eddie,
Slippage was probably a poor choice of words on my part. When loaded the threads will tend to push away from each other and change the ratio of the feed SLIGHTLY. This should not affect your intended usage, but would make it useless for a calibrated lead screw. It will not stop the feed, only change the rate at which it feeds a little bit. Forgive an old man who had too little sleep and not enough coffee when making my original reply.

Related general information for those who might be interested. I don't know of a commercial product using this style of reduction, but there probably is one. There are commercial products that uses the same principle, but a different mechanical layout. One that I am familiar with uses threaded sections of rod around the lead screw. They are retained on the ends in bearings so they provide thrust. The outside rods are opposite handed from the lead screw. IE right hand thread on the lead screw and left hand thread on the planetary rods so the threads mesh. Another actuator looks like a roller bearing except the rollers are set at a slight angle to the axis of rotation. This requires close tolerances in the system as it relies only on pressure between the rollers and the outer housing and shaft to transmit all forces. It can slip in the true sense of the word when loaded but has been used in machine components when a compact high ratio linear force was required, but typically only used limit switches at each end of travel with no requirement for accurate positioning.

This same principle is used on one style of self feeding roller burnisher to finish small bores to very high dimensional accuracy and finish. I have several of them that I use for model engine cylinders. I have been fortunate to be able to buy them used for very reasonable prices as they are quite expensive new. They work wonderful, but it is hard to find the self feeding versions.

Last item, then I will be quiet. Welcome to HMEM Eddie. Please keep us posted on your project. Everyone here is interested in all new things related in any way to engine building. Photos are always welcome.

Gail in NM,USA
Edit: Link to commercial outside roller type added.
http://www.steinmeyer.com/english/kataloge/pdf/drs.pdf

 

[dsquire]

Hi
Been looking around for a while, great site

Quick question
Building a self feeding line-boring bar and I was wondering if
something like this differential screw would work as a feed?
see link:

http://www.youtube.com/watch?v=iPkoGug-oC0&feature=related

I was/ am planning on using an inline speed reducer and a screw with a fine pitch thread (3/4-27) on the end but I came across this
and it looks more compact.
Anyone ever built one?
Any thoughts

Hi Eddie:

I can't say how good this idea is but I have been watching it in another forum for a while. Here is a link to the forum that I have been watching.

A different (better) kind of screw
http://www.cnczone.com/forums/showthread.php?t=13593

I will not even try to explain how it works but you should be able to figure it all out if you check through the whole thread. There is another thread that deals with it also but I am unable to find it right now. When I do I will post a link here for you to check out.

Don