Cutting with a round cutter on the lathe

[professorguy]

I need to put a specific arc into a rod. The radius of the arc must be precise. I created a round cutter with the correct dimensions but this cutter is mounted at 5* to provide relief; the honed top is now an ellipse and the arc is incorrect.



The obvious thing to do is not hone the top of the cutter on an angle, so as to preserve the perfect round cutter geometry (see solution A). But then if that cutter is placed on an angle for relief, what is the shape of the cut? Is is round or elliptical?

To guarantee the geometry, solution B seems the only correct method. But then there is no relief for the cut beyond the curve of the turned material itself. Is that enough?

Let me say that the round cutter needs to cut a single part--its longevity is not an issue. Although my project requires many of these cuts, they are all slightly different and I will make a separate cutter for each one. The cutter is simple O2 steel (hardened and tempered), cutting more O2 steel (untreated).

To get a sense of scale, the diameter of one cutter must be exactly 1.184 mm, one must be exactly 1.136 mm, the rest are slightly smaller. My machinery is marked to hundredths of a millimeter, and I use interpolation by eye for the thousandths.

[fixerdave]

B, but center-drill to the edge and then grind off the backside for chip flow?

David...

Edit... okay, center drill to 1.Xmm might be harsh... still, the idea of making a relief to center...

[Mcgyver]

What is precise mean? you can make a profile cutter by grind and stoning with a radius gauge that'll be fairly accurate, I'd guess easily to within a few thou....but we don't know what precise means. Is there a tolerance, or what's its function?

Using a tangential type setup, go with B. For it it to cut though you need clearance, I've done this by turning a cone so there is clearance then harden/temper.

edit....ok, I see you're talking about working to a 1/1000 of mm. hmmmm....solidly in the good luck department without grinding equipment (do you have a cylindrical grinder?) Even then you'd have to hold your tongue just so. I don't think you are going work to exactly a thousands of a mm by interpolating between .01 graduations....your not going to work to that in any event in a lathe, your spindle bearings have more runout than that.

whats it suppose to do and why does it need to be to three decimal place accuracy (metric)?

[gbritnell]

When you turn up the cutter leave it slightly oversize and cut a 1 degree backward taper from the end. After you harden the cutter grind the top flat until you get the diameter dimension you need.
gbritnell

[Toolguy]

One could machine a piece of tool steel to the correct dimensions with a tapered endmill, then harden the cutter. The cutting tool would be machined upside down, with the taper on the endmill providing the back clearance, giving a form relieved tool. The sizes being discussed are very small. The same could be done with wire EDM. I'm thinking more of a flat piece like a small carbide insert rather than a piece of wire.

[MattiJ]

What is precise mean? you can make a profile cutter by grind and stoning with a radius gauge that'll be fairly accurate, I'd guess easily to within a few thou....but we don't know what precise means. Is there a tolerance, or what's its function?

Using a tangential type setup, go with B. For it it to cut though you need clearance, I've done this by turning a cone so there is clearance then harden/temper.

edit....ok, I see you're talking about working to a 1/1000 of mm. hmmmm....solidly in the good luck department without grinding equipment. I don't think you are going work to exactly a thousands of a mm by interpolating between .01 graduations....your not going to work to a 1/1000mm in any event, your lathe spindle bearings have more runout than that

how about sort of ball turning attachment but instead normal way mount small grinding disk to lathe chuck and put the tool blank to the ballturning-thingy toolholder. Still real bugger even if you have a good DRO in the lathe..

[Magicniner]

When you turn up the cutter leave it slightly oversize and cut a 1 degree backward taper from the end. After you harden the cutter grind the top flat until you get the diameter dimension you need.
gbritnell

+1
Putting the relief on a round cutter allows a the circular cutter to be presented at 90 degrees to the work and create a cut of the required radius.

[Mcgyver]

how about sort of ball turning attachment but instead normal way mount small grinding disk to lathe chuck and put the tool blank to the ballturning-thingy toolholder. Still real bugger even if you have a good DRO in the lathe..

yeah, and there is the boring head approach hold a tangential cutter as well.....but to .001 mm?

[MattiJ]

yeah, and there is the boring head approach hold a tangential cutter as well.....but to .001 mm?

Make ten or fifty and maybe one measures "exactly correct"
Mitutoyo has 0,1um reading micrometer but I image measuring this size of half-ball with mike is a real joy..

[professorguy]

When you turn up the cutter leave it slightly oversize and cut a 1 degree backward taper from the end. After you harden the cutter grind the top flat until you get the diameter dimension you need.

That's the trick I came here to learn! This is brilliant. Since I am making the cutters on the lathe, a reverse taper will be easy to apply. Then I can hone, measure, hone, measure, etc until I get down to the perfect diameter. Thanks Mr. Britnell.

My calculations have 3 significant figures after the decimal, but the actual equipment does not need to be quite so precise--the round cutter is a watchwork wheel cutter-cutter (used to make a wheel cutter). I keep the thou's so I can move the handwheel with confidence, but I recognize that my lathe cannot reproduce at such accuracy. But I think I can get close enough.

[Mcgyver]

keep us posted and hopefully with some pictures.....sounds like interesting work. Around a machine shop, when you say you need exactly.....and give three decimals....it means you need to be that exact. If you can relax it a bit you are off to the races.

George is a very experience machinist, but when I've done this I found it needs a lot more clearance than 1 degree will provide if you are coming in at centre height.

[professorguy]

when I've done this I found it needs a lot more clearance than 1 degree will provide if you are coming in at centre height.

I think I will use a 3*-4* taper just because the taper will only be about 2 mm long, so to provide any significant clearance at all will require a more severe angle than 1*. It will be harder to hone it in to the perfect diameter (since the diameter will change faster with steeper tapers) but the steeper taper will be an easier cut, I should think. The amount of material removed from a 1mm diameter rod tapered only 1* for only 2mm would be almost undetectable.

[boslab]

An inclined plane or section through a round or rod is going to be an ellipse, I suppose, I'm going to guess you would have to cut then recut with a zero rake cutter, or rescrape as it were, if it needs to be an actual radius I'd guess that a milling cutter of the right diameter mounted in a vertical mill and the rod in a rotary device like spin indexer, rotary table or dividing head.
Either would do, or any variant, unless someone has built a tiny ball cutter thing, bet someone has.
Rolling it or grinding it would be the first choice in a shop, a cylynder grinder would make shoes work of it after forming the rad on the wheel.
Mark

[Bob Engelhardt]

...
To get a sense of scale, the diameter of one cutter must be exactly 1.184 mm, one must be exactly 1.136 mm, ...

In machining there is no such thing as "exactly"! It is always a matter of some tolerance. I suppose "1.184" implies +-.0005 (1/2 of .001), but that's moot because you can't machine to that (nothing personal). I would suggest that you find out or figure out what your tolerances actually are. I.e., what they have to be to get parts to fit together, without "too much" slop. If you don't, you will almost certainly wind up with parts that don't fit together.

E.g., if I'm making something that has a shaft in a bored hole & both are dimensioned as 25.000mm, each of them will come out as somewhat different than that, +-. If the shaft is + & the bore -, there's a problem. But if the shaft is dimensioned 25.00 +0,-.01 & the bore 25.01 +.01,-0 you will have a reasonable chance of making them & they will fit.

.001mm is incredibly small. It's 1/25000" or 4/10 of a "tenth". It's the scale where the temperature of the part is important.

Bob

PS - I'm new here & don't know you, so if I've been lecturing what you already know, my apologies.

[Gary Paine]

That's the trick I came here to learn! This is brilliant. Since I am making the cutters on the lathe, a reverse taper will be easy to apply. Then I can hone, measure, hone, measure, etc until I get down to the perfect diameter. Thanks Mr. Britnell.

My calculations have 3 significant figures after the decimal, but the actual equipment does not need to be quite so precise--the round cutter is a watchwork wheel cutter-cutter (used to make a wheel cutter). I keep the thou's so I can move the handwheel with confidence, but I recognize that my lathe cannot reproduce at such accuracy. But I think I can get close enough.

I think you've got it. Also, your cutter made thusly could work if mounted at an angle other than vertical if needed just as long as each cut were taken exactly on the tangent. A vertical mount, though, with the top of the cutter exactly on centerline will allow the crossfeed to work properly to get to depth. You might want to rough out the cut with a different cutter and use the new tool for light finish cuts.