Square peg in a round hole

[Alans Home Workshop]

[DIYer]

Thanks Alans Home Workshop! We've added your Square Stock Holding Method to our Lathes category,
as well as to your builder page: Alans Home Workshop's Homemade Tools. Your receipt:

Square Stock Holding Method by Alans Home Workshop

tags: chuck, holding, collet

[rgsparber]

Given the proper thickness shim, I think I can center square stock in a 3 jaw chuck. I just tried it in my shop. I haven’t derived the equation relating stock size to shim thickness but I’m sure Marv can do it quicker. With one jaw vertical, shim goes on it. Other two jaws press on the stock at the top corners.

My chuck jaws have a taper on the flank so that complicates the math.

[mklotz]

Given the proper thickness shim, I think I can center square stock in a 3 jaw chuck. I just tried it in my shop. I haven’t derived the equation relating stock size to shim thickness but I’m sure Marv can do it quicker. With one jaw vertical, shim goes on it. Other two jaws press on the stock at the top corners.

My chuck jaws have a taper on the flank so that complicates the math.

The ECCENT program on my website* had a program for computing the shim size needed to make eccentrics in a 3jaw. I consider using the 3jaw for jobs that really should be done on a 4jaw an unnecessarily dangerous operation. Even offsetting for an eccentric can be done more safely without a shim using a technique described in the text file that accompanies the program.

A much safer way to handle square work in the 3jaw is to make some cylindrical "collets" as I described here...

Collet inserts for square stock

If you finally break down and mount the 4jaw, you face the problem of centering the stock - a task the "collets" do automatically; this task can be made easier if you add my "flapper" to your DI used for centering...

Centering aid for polygonal stock

--
* The generous soul who hosted my website died on his 50th birthday so the site is no more. However, a fully usable version of it is available on some sort of "way back" machine here...

https://web.archive.org/web/20230526.../mklotz/#intro

[mklotz]

There's perhaps a simpler variant of the square "collets" to which I referred in the previous post.

Take a piece of cylindrical stock and bore it out to be a close sliding fit on your square stock. Slice it end to end with a small end mill so that the 3jaw can clamp down and close it down on the square stock.

If I were doing this (which I wouldn't be), I would arrange for the diameter of the cylindrical stock to be about 10% greater than the diagonal of the square. This so it's flexible enough to close down and grab the square without requiring undue pressure from the 3jaw chuck.

[Bony]

As an interesting obversation worth noting, my regular every day chuck is a 200mm self centering 4 jaw. This type of chuck will hold hexagonal cross section work (eg a nut) which will automatically be self centred. Extraordinary but true, try it for yourself.

[Alans Home Workshop]

Given the proper thickness shim, I think I can center square stock in a 3 jaw chuck. I just tried it in my shop. I haven’t derived the equation relating stock size to shim thickness but I’m sure Marv can do it quicker. With one jaw vertical, shim goes on it. Other two jaws press on the stock at the top corners...

I think the maths for this might be quite complicated

[Alans Home Workshop]

... my "flapper" to your DI used for centering...

Centering aid for polygonal stock

That's a great idea.

[mklotz]

That's a great idea.

Thanks, Alan, I'm glad you like it.

Yes, the edges of polygonal stock are seldom exactly the same distance from the center so better results if you center on the faces rather than the edges. But rotating from face to face puts the sensing stalk of the DI at risk unless you use something like my flapper.

[mklotz]

I think the maths for this might be quite complicated

Accounting for the shape of the jaw tips would make it complex indeed !

If the jaw tips were flat, I think it might be easier. Since the 3jaw is self-centering, the bottom jaw must be at the same distance from the center of the square as the two top jaws are.

Let:

S = length of the side of the square stock

Then S / √2 is the distance of the top jaws from the center of the block and the shim needs to be S / √2 - S/2 = S ( √2 / 2 - 1/2) = S (√2 - 1) / 2 = 0.2071 S

Oh, that it were that simple. :-)

Regardless, shimming isn't nearly as safe as a collet arrangement along the lines of what I described.