Computing offset in 3 jaw lathe chuck

What's even more fun is holding a rectangular section in a three jaw with one jaw a number of scroll threads displaced from the other two. Trial and error, and make the other parts to fit...

I thought that several times (me too stupid). A good number to start with experimenting is: thickness = offset * sqrt(2) [1.41]. But if you have a closer look at your jaws, you'll discover that you don't want to know the formula. The inner contact area of the jaws are concave, have different widths (for different jaws) and if you put a flat in between, things get even more complicated. Well, it could be done, but is it worth spending the time?

You should ask Ignorant, he hosts algebra.com. This way, we'll get rid of him for 2 weeks. :-P

Nick

Yes. A program to calculate the necessary offset is available via the ECCENT archive on my website. The equation is:

p=1.5*e-r+0.5*sqrt(4*r*r-3*e*e+2*e*w*root3-w*w);

where:

p = packing thickness e = eccentric offset r = radius of workpiece w = width of jaws

which can be obtained via straightforward, if a bit messy, trigonometry.

Regards, Marv

Home Shop Freeware - Tools for People Who Build Things

Home Shop Freeware - Tools for People Who Build Things

Seems more logical to build a davit with a threaded attachment and drill and tap a hole in the 4-jaw. Then changing a chuck is a 5 minute project.

Back when I was working in job shops every lathe that had a chuck heavy enough to take two men to lift had a davit attached to it.

Bruce in Bangkok (brucepaigeatgmaildotcom)

I would cheat a bit. Since you know that a .3425 shim gets you a .250 offset, then I would think that 0.001298 per .001 of desired offset would probably do the trick. After everyone goes home, test the theory and if it works out then you can brag how smart you are. If it does not work then you have the option of changing the subject if it comes up again.

One thing I find promising is that your guy asked the question. To me this suggests that he is probably a keeper.

A round part that is chucked in a ideal three jaw chuck is a "circumscribed circle" (a real chuck is more than just a single point). You can find out its radius using this calculator

Well, his name IS Roger!

Actually in looking at the trigonometric solution that iggy posted, that seems to be a viable method.

Marv the equation looks clear except "root3". sqrt(3)???

Hul

Marv wrote:

It still is missing the hard facts. Jaws do have an concave radius. And then it depends wether you put in a round or a flat. The trig by Marv is missing that too.

Nick

Well I guess that it would depend on how close you had to come. If as Tom suggested it was for a cam the degree of precision might not be that important as let's say a cam lobe on an engine.

If you had a bunch of the things to make, I suspect the easy way would be to use a soft collet and make a dead nuts offset, and if you needed one off and needed it to be better than close to eyeball, then you could always put the four jaw on and indicate the part.

Seems to me Tom stumbled on to a shim size that worked for his purpose and If the concave surface on the chuck was a real problem, then I suspect you could perhaps place a shim of given size on jaw 1 and jaw 2 and on jaw 3, use a shim that was the same thickness plus the calculated value. And stand way to the side to be sure that it was snug enough to hold:>)

We use it twice a year, with a davit we might double that.

Most shops looked at it more as a lathe attachment. A post bolted to the floor with a brace to the headstock and the chuck hanging on it but I agree with you that it doesn't get used very frequently.

Bruce in Bangkok (brucepaigeatgmaildotcom)

Yes, root3 is the square root of 3.

My solution is for cylindrical work pieces. I haven't addressed flat workpieces because I can't imagine putting such in the 3jaw. Also, the 'w' in the equation refers to the jaw width of that portion of the jaw that actually contacts the work - not the jaw thickness - so that detail is covered.

Turning offsets this way is an admitted kludge. If your application requires high precision, do it the right way and use the 4jaw.

with "it depends wether you put in a round or a flat", I meant as shim. Sorry for my fuzzyness. So it makes a difference if you shim with a flat or a round.

That was clear.

But it still is concave not flat. :-) I just measured that. On my jaws, there is a concaveness of 0.04mm.

OK, for precise work, using a 4 jaw chuck or an excentric chuck is best, for not so precise work, your formula is certainly good enough.

Ni-nickpicking-ck

replying to Marv, Arthur. wrote: Hi Tom, Very interested in your equation. Could I have a look at your website please? Thanks, Arthur