After thinking about it for a while, doing some maths, and doing some checks, I think I have a method to manually machine an arc on a mil without using a rotary table. I would appreciate it if someone would look to see if I missed something or not: http://benchtopmachineshop.blogspot.com/2013/05/arc-interpolation-on-manual-mill.html

I guess I don't quite follow along with multiplying the half-degree by X or Y... Any chance you could express it as a formula? The standard formula for a cricle centered about the origin (or arbitrary datum) is r[SUP]2[/SUP] = x[SUP]2[/SUP] + y[SUP]2[/SUP] for which x and y can be solved as x = SQRT(r[SUP]2[/SUP] - y[SUP]2[/SUP]) and y = SQRT(r[SUP]2[/SUP] - x[SUP]2[/SUP]). Either way, the issue is granularity. Making discrete adjustments of the hand crank will give you a bunch of zig-zag steps. Also, I'm not sure who your audience is but, someone could come along an nit-pick the daylights out of you over the term "arc" instead of sphere or hemi-sphere in this particual context -unless of course, the metod produces an arc -but I don't think so because at quick glace of the table, the numeric pattern is symmetric. Ray

That is precisely the way a CNC mill works and it will get the job done if the steps are small enough and you have enough patience. I once had to mill a .012" square hole in a piece of brass shim for an optical application. I used a .005" diameter end mill and worked under a microscope to see what was going on. The individual steps were very apparent at that level. Obviously a larger end mill will make the steps less apparent. I think I would get looking for a rotary table though, it's gonna take a lot of time to create an arc with that technique. Tom

I tried a few times and found that its very nerve wracking. . . . just one little "ohs##t" turning the crank the wrong direction and pooof ! "murphy" follows me around most days. . .

Zoltan Assuming a 10.0000 circumference (diameter?) it would seem that I would move 5.000 in the x axis and 5.0000 in the y axis to complete 90 degrees. That would be 180 steps in your spreadsheet. Have I misinterpreted anything? Gene

What type material will the arc be cut into, and how thick will it be? Will the finished arc be on the outside edge of the piece? Or, is it inide of a square piece? As a cabinetmaker, many methods come to mind. Rough out the piece on a band saw. Swing the arc, with a pivot point against an edge sander. I have made very accurate parts that way. A lot depends on the material. Larry

If you're doing 90* of a 10 diameter circle, then yes, you'd move a total of 5 in the X axis and 5 in the Y axis in a total of 180 steps.

I once manually milled my then house numbers 2158 in a fairly fancy font out of 1/2x5 x14" brass. It was time consuming and wasn't quite prefect, but it was pretty close. I pebbled the background inside the border and darkened it with gun blue and polished the foreground and lacquered the works. It can be done. Michael

If you enter in a circumference of 31.415 (the circumference of a circle with diameter 10) then your X (column A) and Y (column B) movement for 90* sums to 5 and 5.

There are all sorts of formulas to describe circles. I just have one question: Is a circle a polygon?

I tried something similar on my 7x12 lathe - the method does work, but you have to careful do it in a way that does not allow for cumulative errors. My lathe does not have DROs, so this was a problem for me. On a mill with DROs I think this method would work just fine, but it takes a long time and it's easy to make a mistake. On the other hand, sometimes you just have to work with what you have, and that's part of what makes hobby machining a fun challenge.

I've made bearing races b4 on the lathe. i turned a piece of stock to the same diameter as my outer race. Then i set up an indicator on my cross slide and had this piece on the ways. Like a tracer.It worked very well, dont see why you couldnt do the same on the mill. These were bearing caps and i was cutting the concave inside of them to fit a radial ball on a shaft. Same as the inside of an outer race on a ball bearing. Never tried it on the mill but im sure it would work. No math involved just hand feeding and keeping your eyes on the indicator

I remember doing a few impromptu jobs, without having access to a rotary table. Hog/Bandsaw, them out to your line, then use a 2-3 foot cheater bar to (feed/finish), the rest of the cuts...only taking off a few thousandths at a time. As long as you don't slip. A slip up could be disastrous. And, a destroyed workpiece. I know some pieces were at least 3/4" thick steel.