how to grind HSS tool bit to 1/8" radius accurately

[metalmagpie]

I want to turn a semicircular cross-section groove in the face of a washer made of 6061-T6 aluminum. The OD of the groove is 2". The washer (disc?) will be held in a 3-jaw chuck with accurately bored soft jaws on a 14x40" lathe. My plan is to grind a tool bit of the right profile and plunge cut after trepanning out a lot of the material first.

I don't have access to an optical comparator, just a set of radius gages. I don't have much confidence in my ability to grind a tool bit to a semicircular profile with any degree of accuracy. I don't think I have any that will work in my junk box of HSS lathe toolbits (don't we all have such a junk box?).

How would you do this grind?

metalmagpie

[JRouche]

Dont suppose you have a ball turner for that lathe. Not saying turn the ball end, but grind it with a dremel type tool post arrangement on the ball turner? Just thinking.. JR

[Paul Alciatore]

Accurately? How accurate? +/-0.010"? +/-0.001"? +/-0.0001"? Or what?

Would a 1/8" round piece of drill rod work if held in a tool holder at 5 degrees to the work surface. That would cut a slightly oval shaped groove, but it would be close to round. 1/8" radius = 1/4" diameter. So use a 1/4" piece of drill rod and grind the end to the rake angle you want and harden it. You can probably use an old or broken 1/4" drill.

If it is held at 5 degrees to the work surface, the error from circular will be 0.0005". That sounds pretty precise to me. Of course, you need to make a holder for it. If that's too much error, five or ten strokes on a stone will take that much off the tip. But then, how are you even going to measure the error? I mean, if the groove is 1/4" wide and 1/8" deep, are you going to measure it at the 45 degree points? How? Who will know?

[mattthemuppet]

Make one of those tangential tool holders and use a 1/8in hss blank with appropriate back rake ground? More left field stuff would be to mill the radius with a external radius carbide router bit? 1/8 ball end mill with one flute ground off - cut one side in forward then rotate the end mill 180 deg and run the lathe in reverse to cut the other side of the groove?

[mattthemuppet]

Bah, too slow

[J Tiers]

I have ground a fly cutter to a gear profile, I would think a round would be easier.

GRind to width, longer than you really need. Grind to as good a round as you can, by eye or whatever. Now try your gauge against it (I used a similar gear), and wherever you spot excess material, grind off a bit. Keep trying it against the gauge and touching up untl you get a good fit.

You can do the last bit by hand with an Arkansas stone or similar. Hold gauge and cutter up to the light, and where the gauge shows the light blocked, grind off until there is no more light area, so it is all blocked.

You can do quite well that way. And with any shape that you have the mating part to. It's not really different to what you would do with a comparator, which would just show you the true shape vs what you have.

If it comes up a lot, you could make a fixture. But the nature of the process is such that the effective pivot point must be "inside" the wheel. So the frame has to surround the wheel somewhat. Could be biggish.

[elf]

I recently had to make sockets to fit 14mm ball bearings. I glued the toolbit to a sacrificial block that had a pivot hole 7mm from the front. This was placed on another sacrificial block that had a 3mm pin inserted in it. The toolrest was set at 90 degrees to the wheel. The toolbit/sacrificial block was rotated as the bottom sacrificial block was advanced into the wheel. When the front of the toolbit/sacrificial block touched the wheel, the toolrest was rotated 10 degrees to provide relief. I used diamond hones to finish the relief as the shape when grinding the relief isn't a true hemisphere. I couldn't detect any slop in the ball joints. It all depends on how accurately you can position the pivot hole.

[BCRider]

You say you don't trust your ability NOW. But you've got a radius gauge set to work with? In that case find a good strong magnifying glass and use a back light when doing the comparison to make small errors jump out more clearly.

An excellent magnifying glass with very little distortion is the 50mm lens of an old 35mm film SLR camera. With that lens held in an arm and using the back lighting idea you will see all but the smallest errors.

As for the skill you say you don't have? Well start out with trying and re-try. After a half dozen pokes at the grinding wheel that skill you're worried about not having will be growing more and more.

And may I suggest that this would be a good time to do the grinding on the round edge of a normal grinding wheel and not the flat face of a cup stone? That way smaller errors can be dressed out with a few swipes across a sharpening stone instead of hogging them off with the grinding wheel. Using the round edge will give the cutter a hollow ground relief which makes it easier to stone away the amount you require at the cutting edge.

[Ian B]

There are these things: http://www.ebay.co.uk/itm/Manchester...oAAOSwv0tVbiPb which should be available in the radius that you need. Otherwise, how about milling it with a 1/4" ball end mill and a rotary table?

As it's aluminium, you could finish the profile using several 1/4" steel balls from a ball bearing, set the work up like a thrust bearing with a hardened thrust bearing ring on the other side. They'll burnish the surface for you and the depth will be as accurate as the hardened thrust ring.

Ian

[Tony Ennis]

Accurately? How accurate? +/-0.010"? +/-0.001"? +/-0.0001"?

This is pretty important to know. Removing the corners by grinding at 45, then the touching these corners again kinda gets you there.

[JCHannum]

1/8" radius is 1/4" diameter. A slice of 1/4" OD HSS, carbide or drill rod silver soldered or clamped to a piece of square stock is simple enough to gin up if nothing else is available.

[Ian B]

The only real difficulty here is that the OP wants to cut this groove in the face, rather than the periphery of the work. Clearance on the outer face of the groove then becomes an issue, which is hard to deal with when using any kind of tangential tool. It's a fairly small diameter - the groove ends up 2" OD, 1.5" ID. JC's slice would work, with a bit of relief.

Brazing a hardened 1/4" ball onto a round rod and then grinding the top half off the ball would give a reasonable result, especially as it's aluminium and it's a one-off.

Ian

[mars-red]

If it was me, I'd do what JC suggests and make a button of the correct diameter. I wouldn't even bother with HSS, I use O1 for stuff like this all the time.

[Dan Dubeau]

It's aluminum, 1 part. Drill a piece of steel for a "toolbit", Cut down to the quadrants, finish the rad with a file, add a bit of lead at the corners, and cut your one part. Don't even bother to harden it. I'd knock the corner of the part off a bit first so you're not doing a full pass.

Or you could practice your grinding a bit, and do it out of HSS. It's not hard to learn, and for that small of a profile you'd be done pretty quick.

[LKeithR]

...There are these things: http://www.ebay.co.uk/itm/Manchester...oAAOSwv0tVbiPb which should be available in the radius that you need....

Yeah, there are many inserts available with a pre-ground radius. Can't imagine why you'd want to mess around grinding something...